Shared genetic factors between osteoarthritis and cardiovascular disease may underlie common etiology.

Osteoarthritis is one of the most common musculoskeletal diseases and increases the risk of severe cardiovascular disease, like heart attack and stroke. In some individuals, osteoarthritis and cardiovascular disease will co-occur. This co-occurrence might be due to shared risk factors, for example high age, lifestyle factors and/or a shared genetic liability for the two diseases. Here, we show that the correlation between osteoarthritis and cardiovascular disease can be explained by shared genetic factors, independent of high age and body weight, and also likely independent of lifestyle factors, like smoking and physical activity level. Findings suggest that genetic factors that are shared for osteoarthritis and cardiovascular disease may contribute to both diseases. Thus, the prevailing idea that osteoarthritis is predominantly a risk factor for cardiovascular disease is challenged. Our findings imply that the current diagnostic boundaries between these diseases may need to be re-evaluated.

Arthroscopic partial meniscectomy versus exercise therapy for degenerative meniscal tears: 10-year follow-up of the OMEX randomised controlled trial.

OBJECTIVE: To evaluate radiographic knee osteoarthritis (OA) progression, development of knee OA, patient-reported outcomes and knee muscle strength at 10-year follow-up after arthroscopic partial meniscectomy (APM) or exercise therapy for degenerative meniscal tears. METHODS: Randomised controlled trial including 140 participants, with a degenerative meniscal tear and no or minimal radiographic OA changes. Participants were randomised to either APM or 12 weeks of exercise therapy (1:1 ratio). The primary outcome was knee OA progression assessed by the Osteoarthritis Research Society International (OARSI) atlas sum score (sum of medial and lateral compartment joint space narrowing and osteophyte score). Secondary outcomes included incidence of radiographic and symptomatic knee OA, patient-reported pain and knee function and isokinetic knee muscle strength. RESULTS: The adjusted mean difference in change in the OARSI sum score was 0.39 (95% CI -0.19 to 0.97), with more progression in the APM group. The incidence of radiographic knee OA was 23% in the APM group and 20% in the exercise group (adjusted risk difference 3% (95% CI -13% to 19%)). No clinically relevant differences were found in patient-reported outcomes or isokinetic knee muscle strength. CONCLUSION: No differences in radiographic knee OA progression and comparable rates of knee OA development were observed 10 years following APM and exercise therapy for degenerative meniscal tears. Both treatments were associated with improved patient-reported pain and knee function. TRIAL REGISTRATION NUMBER: NCT01002794.

Effect of arthroscopic partial meniscectomy on structural degeneration of the knee - A 5-year MRI-based follow-up of the placebo-surgery controlled FIDELITY (Finnish Degenerative Meniscus Lesion Study) trial.

OBJECTIVE: To assess the 5-year effects of arthroscopic partial meniscectomy (APM) vs. placebo surgery on the development of the structural changes of the knee by magnetic resonance imaging (MRI). DESIGN: This multicentre, randomized, participant- and outcome-assessor-blinded, placebo-surgery-controlled trial was carried out in Finland. We randomized 146 adults, mean age 52 years (range 35 to 65) to undergo either APM or placebo surgery. The subjects had symptoms of degenerative medial meniscus tear, a tear verified in MRI and arthroscopy, and no advanced osteoarthritis at baseline. We compared the baseline and 5-year follow-up MRIs using MRI Osteoarthritis Knee Score scoring to derive subregional data on cartilage damage, osteophytes and bone marrow lesions (BMLs). Progression of structural cartilage changes analyzed per subregion was the main outcome, that of osteophytes and BMLs secondary outcomes. We analyzed the progression with multilevel logistic regression model on subregion-level data, adjusted for randomization stratification factors, and using robust standard errors. RESULTS: Sixty-three (90%) subjects in the APM and 73 (96%) in the placebo-surgery group had MRI at both time points. The adjusted odds ratio (APM vs. placebo-surgery) was 1.31 (95% confidence interval 0.81, 1.94) for progression of cartilage damage, 2.86 (1.16, 6.21) for osteophytes, and 1.43 (0.84, 2.43) for BMLs. CONCLUSIONS: We found a slightly greater risk for progression of osteophytes in the APM group compared to the placebo-surgery group at 5 years after surgery. TRIAL REGISTRATION: ClinicalTrials.gov (NCT01052233 and NCT00549172).

Prevalence and classification of meniscal calcifications in the human knee.

OBJECTIVE: To investigate the occurrence of meniscal calcifications in individuals with and without knee osteoarthritis (OA). Additionally, we aim to identify the specific types of calcifications: basic calcium phosphate (BCP) and calcium pyrophosphate dihydrate (CPP). METHOD: We analyzed 82 meniscal posterior horn samples (medial and lateral) collected from 41 human subjects. Among them, 20 individuals underwent total knee replacement due to medial compartment OA, while 21 deceased donors had no known knee OA. The assessment of meniscal calcifications and Pauli's histopathological scoring was conducted using histological sections. Furthermore, adjacent sections underwent measurement using Raman spectroscopy to characterize BCP and CPP calcifications based on their distinct spectral fingerprints. RESULTS: All OA individuals exhibited calcifications in at least one meniscus, compared to 9.5% (95%CI 1%, 30%) of donors. Among 35 OA menisci with calcifications, 28(80%) had BCP, 5(14%) had CPP and 2(6%) had both types. In 4 donor menisci, 3(75%) had CPP while 1(25%) had both types. We estimated the association between Pauli score and presence of BCP in OA individuals, yielding an odds ratio of 2.1 (95%CI 0.8, 5.3) per 1 Pauli score. The association between Pauli score and presence of CPP (in whole study sample) seemed weaker, with odds ratio of 1.3 (95%CI 1.1, 1.7). CONCLUSION: The presence of BCP was predominant in menisci of OA individuals, whereas CPP exhibited similar prevalence in individuals with and without OA. The formation of BCP crystals in menisci may represent an important and specific characteristic of OA disease process that warrants further attention.

Uncovering sociodemographic disparities in temporal trends of osteoarthritis incidence and age-at-diagnosis, 2006-2019.

AIM: To describe sociodemographic disparities in temporal trends of incidence and age distributions of first registered osteoarthritis (OA) diagnosis in southern Sweden. METHODS: We identified all Skåne residents aged 35+ who had lived in the region at any point during the period 2006-2019 with no previous OA diagnosis (ICD-10 codes M15-M19) for 8 years prior to inclusion in the study (n = 849,061). We calculated person-years from inclusion until OA diagnosis, death, emigration, or 31 December 2019, whichever occurred first. Combining sex (female, male), education (low, medium, high) and nativity (Swedish, immigrant), we created a variable with 12 strata. Average annual percent changes in age-standardized incidence rates were estimated using joinpoint regression. Changes in the median age-at-diagnosis (year of diagnosis minus birth year), weighted to the mid-2005 Swedish population, were explored. RESULTS: Cumulative age-standardized incidence rates ranged from 116 (95% CI: 111, 121) per 10,000 person-years for immigrant males with low education to 205 (95% CI: 200, 210) for immigrant females with medium education. The estimated average annual percent changes (ranging from 3.4% to 6.1%) were generally similar, with slightly greater variations among immigrants than Swedes. The weighted median age-at-diagnosis was higher for Swedes and low educated people. Immigrant females with low education were the only stratum with a reduction (3 years) in the weighted median age-at-diagnosis over time. Sociodemographic patterns in knee OA incidence were different from patterns for hip OA. CONCLUSIONS: There were few sociodemographic disparities in temporal trends of OA incidence and age-at-diagnosis, suggesting persistent sociodemographic disparities in OA burden in southern Sweden.

Rheumatic and Musculoskeletal Diseases and Risk of Dementia: A Nested Case-Control Study.

OBJECTIVE: To investigate the associations between rheumatic and musculoskeletal diseases (RMDs) and incident dementia using population register-based data. METHODS: This nested case-control study was conducted based on a cohort of residents in the Skåne region, Sweden, aged 50 years and older in 2009 without doctor-diagnosed dementia during 1998 to 2009 (n = 402,825). Individuals with a new main diagnosis of dementia during 2010 to 2019 were identified as incident patients with dementia (n = 22,131). Controls without diagnosed dementia were randomly matched 1:1 by sex, age, and Elixhauser comorbidity index using incidence density sampling. Separate conditional logistic regression analyses adjusted for confounders were fitted for the following RMDs, diagnosed at least 2 years before dementia diagnosis as exposure: gout, osteoarthritis, rheumatoid arthritis, spondyloarthropathies (SpA), and systemic connective tissue disorders. Subgroup analyses by dementia subtype, sex, age, comorbidity, and RMDs/dementia identification were conducted. RESULTS: Although gout (adjusted rate ratio 0.88; 95% confidence interval 0.79-0.97), osteoarthritis (0.92; 0.88-0.96), and systemic connective tissue disorders (0.91; 0.83-0.99) were associated with decreased risk of dementia, the associations for rheumatoid arthritis (1.05; 0.92-1.19) and SpA (1.17; 0.94-1.45) were inconclusive. The associations between RMDs and incident dementia were similar across sex, age, and comorbidity subgroups with a few exceptions (eg, an adjusted rate ratio of 0.99 [95% confidence interval 0.71-1.39] in males vs 1.31 [0.99-1.74] in female patients for SpA). CONCLUSION: Persons with diagnosed RMDs seem to have comparable or slightly lower risks of developing dementia compared with those without known RMD.

Frequent use of prescription NSAIDs among people with knee or hip osteoarthritis despite contraindications to or precautions with NSAIDs.

OBJECTIVE: To describe the use of non-steroidal anti-inflammatory drugs (NSAID), opioids, and physiotherapy (PT) among persons with newly diagnosed knee or hip osteoarthritis (OA) with and without NSAID contraindications or precautions. DESIGN: We used population-based register data to identify adults aged ≥35 as of January 1, 2014, residing in Skåne region (Sweden) between 2004 and 2013, without a previous knee or hip OA diagnosis. Among this cohort, we identified people with incident knee or hip OA diagnosis between 2014 and 2018 and the presence of contraindications to or precautions for oral NSAIDs at the time of OA diagnosis. We estimated the risk of 1) regular oral NSAID use, 2) regular opioid use, and 3) PT during the first year after diagnosis among those with vs. without contraindications or precautions using confounder-adjusted logistic regression with standardization. RESULTS: We identified 35,173 persons with newly diagnosed OA, of whom 3257 and 8351 had ≥1 contraindication to oral NSAIDs and ≥1 precaution, respectively. Overall, 27% of individuals used oral NSAIDs (with or without opioids or PT), 10% used opioids, and 57% attended PT. Among patients with contraindications, 21% used oral NSAIDs compared to 31% without (absolute adjusted difference -0.06 (95% CIs: -0.08, -0.05)), 53% vs 59% used PT (adjusted difference -0.03 (-0.05, -0.01)), while 14% vs. 8% had prescribed dispensed opioids (adjusted difference 0.02 (0.01, 0.03)). Similar results were observed for those with precautions. CONCLUSIONS: We highlight the need for safer treatment options. People with OA and contraindications/precautions to NSAIDs have a higher risk of opioid use, slightly lower risk of PT use, and continue to be prescribed NSAIDs.

Twenty-year trajectories of morbidity in individuals with and without osteoarthritis.

OBJECTIVES: To identify multimorbidity trajectories over 20 years among incident osteoarthritis (OA) individuals and OA-free matched references. METHODS: Cohort study using prospectively collected healthcare data from the Skåne region, Sweden (~1.4 million residents). We extracted diagnoses for OA and 67 common chronic conditions. We included individuals aged 40+ years on 31 December 2007, with incident OA between 2008 and 2009. We selected references without OA, matched on birth year, sex, and year of death or moving outside the region. We employed group-based trajectory modelling to capture morbidity count trajectories from 1998 to 2019. Individuals without any comorbidity were included as a reference group but were not included in the model. RESULTS: We identified 9846 OA cases (mean age: 65.9 (SD 11.7), female: 58%) and 9846 matched references. Among both cases and references, 1296 individuals did not develop chronic conditions (no-chronic-condition class). We identified four classes. At the study outset, all classes exhibited a low average number of chronic conditions (≤1). Class 1 had the slowest progression towards multimorbidity, which increased progressively in each class. Class 1 had the lowest count of chronic conditions at the end of the follow-up (mean: 2.9 (SD 1.7)), while class 4 had the highest (9.6 (2.6)). The presence of OA was associated with a 1.29 (1.12, 1.48) adjusted relative risk of belonging to class 1 up to 2.45 (2.12, 2.83) for class 4. CONCLUSIONS: Our findings suggest that individuals with OA face an almost threefold higher risk of developing severe multimorbidity.

Arthroscopic Partial Meniscectomy for a Degenerative Meniscus Tear Is Not Cost Effective Compared With Placebo Surgery: An Economic Evaluation Based on the FIDELITY Trial Data.

BACKGROUND: In patients with a degenerative tear of the medial meniscus, recent meta-analyses and systematic reviews have shown no treatment benefit of arthroscopic partial meniscectomy (APM) over conservative treatment or placebo surgery. Yet, advocates of APM still argue that APM is cost effective. Giving advocates of APM their due, we note that there is evidence from the treatment of other musculoskeletal complaints to suggest that a treatment may prove cost effective even in the absence of improvements in efficacy outcomes, as it may lead to other benefits, such as diminished productivity loss and reduced costs, and so the question of cost effectiveness needs to be answered for APM. QUESTIONS/PURPOSES: (1) Does APM result in lower postoperative costs compared with placebo surgery? (2) Is APM cost-effective compared with placebo surgery? METHODS: One hundred forty-six adults aged 35 to 65 years with knee symptoms consistent with a degenerative medial meniscus tear and no knee osteoarthritis according to the American College of Rheumatology clinical criteria were randomized to APM (n = 70) or placebo surgery (n = 76). In the APM and placebo surgery groups, mean age was 52 ± 7 years and 52 ± 7 years, and 60% (42 of 70) and 62% (47 of 76) of participants were men, respectively. There were no between-group differences in baseline characteristics. In both groups, a standard diagnostic arthroscopy was first performed. Thereafter, in the APM group, the torn meniscus was trimmed to solid meniscus tissue, whereas in the placebo surgery group, APM was carefully mimicked but no resection of meniscal tissue was performed; as such, surgical costs were the same in both arms and were not included in the analyses. All patients received identical postoperative care including a graduated home-based exercise program. At the 2-year follow-up, two patients were lost to follow-up, both in the placebo surgery group. Cost effectiveness over the 2-year trial period was computed as incremental net monetary benefit (INMB) for improvements in quality-adjusted life years (QALY), using both the societal (primary) and healthcare system (secondary) perspectives. To be able to consider APM cost effective, the CEA analysis should yield a positive INMB value. Nonparametric bootstrapping was used to assess uncertainty. Several one-way sensitivity analyses were also performed. RESULTS: APM did not deliver lower postoperative costs, nor did it convincingly improve quality of life scores when compared with placebo surgery. From a societal perspective, APM was associated with € 971 (95% CI -2013 to 4017) higher costs and 0.015 (95% CI -0.011 to 0.041) improved QALYs over 2-year follow-up compared with placebo surgery. Both differences were statistically inconclusive (a wide 95% CI that crossed the line of no difference). Using the conventional willingness to pay (WTP) threshold of € 35,000 per QALY, APM resulted in a negative INMB of € -460 (95% CI -3757 to 2698). In our analysis, APM would result in a positive INMB only when the WTP threshold rises to about € 65,000 per QALY. The wide 95% CIs suggests uncertain cost effectiveness irrespective of chosen WTP threshold. CONCLUSION: The results of this study lend further support to clinical practice guidelines recommending against the use of APM in patients with a degenerative meniscus tear. Given the robustness of existing evidence demonstrating no benefit or cost effectiveness of APM over nonsurgical treatment or placebo surgery, future research is unlikely to alter this conclusion.Level of Evidence Level III, economic analysis.

The coexistence of diabetes, hypertension and obesity is associated with worse pain outcomes following exercise for osteoarthritis: A cohort study on 80,893 patients.

OBJECTIVES: To investigate how the co-occurrence of diabetes, hypertension and overweight/obesity is associated with pain following an exercise intervention for knee and hip osteoarthritis (OA). METHODS: Register-based cohort study. We included people from the Swedish Osteoarthritis Register who underwent education and exercise for knee or hip OA. Diabetes and hypertension were defined using medical records and dispensation of medication. Body Mass Index (BMI) was used to identify people with overweight (≥25 to <30), and obesity (≥30). We used linear mixed-effect models with patients nested into clinics to estimate the associations between the exposures and pain (Numeric Rating Scale 0-10), adjusting for age, sex, education, and physical activity. RESULTS: We analysed 80,893 patients with knee or hip OA. The accumulation of metabolic conditions was associated with worse pain at baseline and follow-ups. When obesity, hypertension and diabetes coexisted, patients treated for knee OA reported more pain at baseline (adjusted mean pain difference 0.9 [95 %CI: 0.8; 1.0]), 3 months (1.0 [0.9; 1.1]) and 12 months (1.3 [1.1; 1.4]) compared to those without any of the conditions. Similar results were observed for patients treated for hip OA when obesity, hypertension and diabetes coexisted (baseline (0.7 [0.5; 0.8], 3 (0.8[0.6; 1.0]) and 12 months (1.1[0.8; 1.3]). CONCLUSIONS: When diabetes, hypertension and obesity coexist with OA, patients not only experience heightened baseline pain compared to metabolically healthy individuals, but the disparity increases after an education and exercise intervention suggesting that a one-size-fits-all approach may be inadequate in addressing the complex interplay between metabolic health and OA.

Exploring the Early Molecular Pathogenesis of Osteoarthritis Using Differential Network Analysis of Human Synovial Fluid.

The molecular mechanisms that drive the onset and development of osteoarthritis (OA) remain largely unknown. In this exploratory study, we used a proteomic platform (SOMAscan assay) to measure the relative abundance of more than 6000 proteins in synovial fluid (SF) from knees of human donors with healthy or mildly degenerated tissues, and knees with late-stage OA from patients undergoing knee replacement surgery. Using a linear mixed effects model, we estimated the differential abundance of 6251 proteins between the three groups. We found 583 proteins upregulated in the late-stage OA, including MMP1, collagenase 3 and interleukin-6. Further, we selected 760 proteins (800 aptamers) based on absolute fold changes between the healthy and mild degeneration groups. To those, we applied Gaussian Graphical Models (GGMs) to analyze the conditional dependence of proteins and to identify key proteins and subnetworks involved in early OA pathogenesis. After regularization and stability selection, we identified 102 proteins involved in GGM networks. Notably, network complexity was lost in the protein graph for mild degeneration when compared to controls, suggesting a disruption in the regular protein interplay. Furthermore, among our main findings were several downregulated (in mild degeneration versus healthy) proteins with unique interactions in the healthy group, one of which, SLCO5A1, has not previously been associated with OA. Our results suggest that this protein is important for healthy joint function. Further, our data suggests that SF proteomics, combined with GGMs, can reveal novel insights into the molecular pathogenesis and identification of biomarker candidates for early-stage OA.

2023 EULAR classification criteria for hand osteoarthritis.

OBJECTIVES: The objective of this study is to develop classification criteria for overall hand osteoarthritis (OA), interphalangeal OA and thumb base OA based on self-reported data and radiographic features. METHODS: The classification criteria sets were developed in three phases. In phase 1, we identified criteria that discriminated hand OA from controls. In phase 2, we used a consensus-based decision analysis approach to derive a clinician-based evaluation of the relative importance of the criteria. In phase 3, we refined the scoring system, determined the cut-offs for disease classification and compared the sensitivity and specificity of the European Alliance of Associations for Rheumatology (EULAR) criteria with the 1990 American College of Rheumatology (ACR) criteria. RESULTS: In persons with hand symptoms and no other disease (including psoriasis) or acute injury that can explain the hand symptoms (mandatory criteria), hand OA can be classified based on age, duration of morning stiffness, number of joints with osteophytes and joint space narrowing, and concordance between symptoms and radiographic findings. Using a sum of scores based on each diagnostic element, overall hand OA can be classified if a person achieves 9 or more points on a 0-15 scale. The cut-off for interphalangeal OA and thumb base OA is 8 points. While the EULAR criteria demonstrated better sensitivity than the ACR criteria in the phase 1 data set, the performance of the two criteria sets was similar in two external cohorts. CONCLUSIONS: International experts developed the EULAR criteria to classify overall hand OA, interphalangeal OA and thumb base OA in clinical studies using a rigorous methodology.

Myocardial infarction in a population-based cohort of patients with biopsy-confirmed giant cell arteritis in southern Sweden.

OBJECTIVES: To determine the incidence rate (IR) of myocardial infarction (MI), relative risk of MI, and impact of incident MI on mortality in individuals with biopsy-confirmed giant cell arteritis (GCA). METHODS: MIs in individuals diagnosed with GCA 1998-2016 in Skåne, Sweden were identified by searching the SWEDEHEART register, a record of all patients receiving care for MI in a coronary care unit (CCU). The regional diagnosis database, with subsequent case review, identified GCA patients receiving care for MI outside of a CCU. A cohort of 10 reference subjects for each GCA case, matched for age, sex and area of residence, was used to calculate the incidence rate ratio (IRR) of MI in GCA to that in the general population. RESULTS: The GCA cohort comprised 1134 individuals. During 7958 person-years of follow-up, 102 were diagnosed with incident MI, yielding an IR of 12.8 per 1000 person-years (95% CI 10.3 to 15.3). The IR was highest in the 30 days following GCA diagnosis and declined thereafter. The IRR of MI in GCA to that of the background population was 1.29 (95% CI 1.05 to 1.59). Mortality was higher in GCA patients who experienced incident MI than in those without MI (HR 2.8; 95% CI 2.2 to 3.6). CONCLUSIONS: The highest incidence of MI occurs within the 30 days following diagnosis of GCA. Individuals with GCA have a moderately increased risk of MI compared with a reference population. Incident MI has a major impact on mortality in GCA.

Changes in educational inequalities in knee and hip osteoarthritis surgery and non-surgery specialist care visits over time in Sweden.

Objective: To examine changes in prevalence and socioeconomic inequalities in knee and hip OA outcomes, in more specific surgery and non-surgery specialist care visits, from 2001 to 2011 in Sweden and to what extent sociodemographic factors can explain the changes. Design: We included all individuals aged ≥35 years resident in Sweden from 2001 to 2011. Individual-level data was retrieved from the Swedish Interdisciplinary Panel. Highest educational attainment was used as socioeconomic measure and the concentration index was used to assess relative and absolute educational inequalities. We used decomposition method to examine changes in prevalence and relative educational inequalities. Results: A total of 4,794,693 and 5,359,186 people were included for the years 2001 and 2011, respectively. The crude prevalence of surgery and specialist visits for knee and hip OA was 36-83% higher in 2011 than in 2001. The increase in hip OA outcomes was largely explained by changes in the sociodemographic composition of the population, whereas for knee OA outcomes, changes in the strength of the associations with sociodemographic factors appeared more important. All outcomes were concentrated among people with lower education in all study years. The relative inequalities declined over the study period, while the absolute inequalities increased for knee OA outcomes and remained stable for hip OA. Conclusion: Our findings show an increasing burden of all studied OA outcomes. Moreover, our findings suggest persistent educational inequalities with more surgeries and specialist visits among lower-educated individuals. Future research should incorporate additional variables to better understand and address these inequalities.

Deep Learning for Predicting Progression of Patellofemoral Osteoarthritis Based on Lateral Knee Radiographs, Demographic Data, and Symptomatic Assessments.

OBJECTIVE: In this study, we propose a novel framework that utilizes deep learning and attention mechanisms to predict the radiographic progression of patellofemoral osteoarthritis (PFOA) over a period of 7 years. MATERIAL AND METHODS: This study included subjects (1,832 subjects, 3,276 knees) from the baseline of the Multicenter Osteoarthritis Study (MOST). Patellofemoral joint regions of interest were identified using an automated landmark detection tool (BoneFinder) on lateral knee X-rays. An end-to-end deep learning method was developed for predicting PFOA progression based on imaging data in a five-fold cross-validation setting. To evaluate the performance of the models, a set of baselines based on known risk factors were developed and analyzed using gradient boosting machine (GBM). Risk factors included age, sex, body mass index, and Western Ontario and McMaster Universities Arthritis Index score, and the radiographic osteoarthritis stage of the tibiofemoral joint (Kellgren and Lawrence [KL] score). Finally, to increase predictive power, we trained an ensemble model using both imaging and clinical data. RESULTS: Among the individual models, the performance of our deep convolutional neural network attention model achieved the best performance with an area under the receiver operating characteristic curve (AUC) of 0.856 and average precision (AP) of 0.431, slightly outperforming the deep learning approach without attention (AUC = 0.832, AP = 0.4) and the best performing reference GBM model (AUC = 0.767, AP = 0.334). The inclusion of imaging data and clinical variables in an ensemble model allowed statistically more powerful prediction of PFOA progression (AUC = 0.865, AP = 0.447), although the clinical significance of this minor performance gain remains unknown. The spatial attention module improved the predictive performance of the backbone model, and the visual interpretation of attention maps focused on the joint space and the regions where osteophytes typically occur. CONCLUSION: This study demonstrated the potential of machine learning models to predict the progression of PFOA using imaging and clinical variables. These models could be used to identify patients who are at high risk of progression and prioritize them for new treatments. However, even though the accuracy of the models were excellent in this study using the MOST dataset, they should be still validated using external patient cohorts in the future.

Toward designing human intervention studies to prevent osteoarthritis after knee injury: A report from an interdisciplinary OARSI 2023 workshop.

Objective: The global impact of osteoarthritis is growing. Currently no disease modifying osteoarthritis drugs/therapies exist, increasing the need for preventative strategies. Knee injuries have a high prevalence, distinct onset, and strong independent association with post-traumatic osteoarthritis (PTOA). Numerous groups are embarking upon research that will culminate in clinical trials to assess the effect of interventions to prevent knee PTOA despite challenges and lack of consensus about trial design in this population. Our objectives were to improve awareness of knee PTOA prevention trial design and discuss state-of-the art methods to address the unique opportunities and challenges of these studies. Design: An international interdisciplinary group developed a workshop, hosted at the 2023 Osteoarthritis Research Society International Congress. Here we summarize the workshop content and outputs, with the goal of moving the field of PTOA prevention trial design forward. Results: Workshop highlights included discussions about target population (considering risk, homogeneity, and possibility of modifying osteoarthritis outcome); target treatment (considering delivery, timing, feasibility and effectiveness); comparators (usual care, placebo), and primary symptomatic outcomes considering surrogates and the importance of knee function and symptoms other than pain to this population. Conclusions: Opportunities to test multimodal PTOA prevention interventions across preclinical models and clinical trials exist. As improving symptomatic outcomes aligns with patient and regulator priorities, co-primary symptomatic (single or aggregate/multidimensional outcome considering function and symptoms beyond pain) and structural/physiological outcomes may be appropriate for these trials. To ensure PTOA prevention trials are relevant and acceptable to all stakeholders, future research should address critical knowledge gaps and challenges.

Exploratory neutron tomography of articular cartilage.

OBJECTIVE: To investigate the feasibility of using neutron tomography to gain new knowledge of human articular cartilage degeneration in osteoarthritis (OA). Different sample preparation techniques were evaluated to identify maximum intra-tissue contrast. DESIGN: Human articular cartilage samples from 14 deceased donors (18-75 years, 9 males, 5 females) and 4 patients undergoing total knee replacement due to known OA (all female, 61-75 years) were prepared using different techniques: control in saline, treated with heavy water saline, fixed and treated in heavy water saline, and fixed and dehydrated with ethanol. Neutron tomographic imaging (isotropic voxel sizes from 7.5 to 13.5 µm) was performed at two large scale facilities. The 3D images were evaluated for gradients in hydrogen attenuation as well as compared to images from absorption X-ray tomography, magnetic resonance imaging, and histology. RESULTS: Cartilage was distinguishable from background and other tissues in neutron tomographs. Intra-tissue contrast was highest in heavy water-treated samples, which showed a clear gradient from the cartilage surface to the bone interface. Increased neutron flux or exposure time improved image quality but did not affect the ability to detect gradients. Samples from older donors showed high variation in gradient profile, especially from donors with known OA. CONCLUSIONS: Neutron tomography is a viable technique for specialized studies of cartilage, particularly for quantifying properties relating to the hydrogen density of the tissue matrix or water movement in the tissue.

Exercise for knee osteoarthritis pain: Association or causation?

Exercise is universally recommended as a primary strategy for the management of knee osteoarthritis (OA) pain. The recommendations are based on results from more than 100 randomized controlled trials (RCTs) that compare exercise to no-attention control groups. However, due to the inherent difficulties with adequate placebo control, participant blinding and the use of patient-reported outcomes, the existing RCT evidence is imperfect. To better understand the evidence used to support a causal relationship between exercise and knee OA pain relief, we examined the existing evidence through the Bradford Hill considerations for causation. The Bradford Hill considerations, first proposed in 1965 by Sir Austin Bradford Hill, provide a framework for assessment of possible causal relationships. There are 9 considerations by which the evidence is reviewed: Strength of association, Consistency, Specificity, Temporality, Biological Gradient (Dose-Response), Plausibility, Coherence, Experiment, and Analogy. Viewing the evidence from these 9 viewpoints did neither bring forward indisputable evidence for nor against the causal relationship between exercise and improved knee OA pain. Rather, we conclude that the current evidence is not sufficient to support claims about (lack of) causality. With our review, we hope to advance the continued global conversation about how to improve the evidence-based management of patients with knee OA.

Genetic Influence on Osteoarthritis Versus Other Rheumatic Diseases.

OBJECTIVE: We aimed to compare the genetic contribution to osteoarthritis (OA) versus other rheumatic/musculoskeletal diseases (RMDs) in the same population and to explore the role for any shared genetics between OA and other RMDs. METHODS: In 59,970 Swedish twins aged 35 years or older, we estimated the heritability (the variance explained by genetic factors) of OA in peripheral joints, back and neck pain, shoulder pain (adhesive capsulitis, impingement syndrome, etc), rheumatoid arthritis, spondyloarthritis (SpA) and psoriatic arthritis, myalgia, and osteoporosis diagnosed in specialist and inpatient care. We also studied how much covariance between OA and each of the RMDs could be explained by genetics by studying phenotypic correlations in bivariate classical twin models. RESULTS: Any-site OA and hip OA (50% and 64%) were among the most heritable RMDs (as compared with 23% for fibromyalgia [lowest] and 63% for SpA [highest]). The highest phenotypic correlations were between OA (any joint site) and shoulder pain in the same individual (r = 0.33, 95% confidence interval 0.31-0.35), of which 70% (95% confidence interval 52-88) could be explained by shared genetics. The phenotypic correlation between OA and back/neck pain was r = 0.25, with 25% to 75% explained by genetics. Phenotypic correlations between OA and each of the other RMDs were lower (r ~ 0.1 to r ~ 0.2), with inconclusive sources of variation. CONCLUSION: OA has relatively large heritability as compared with other RMDs. The coexistence of OA and shoulder pain, as well as back pain, was common and could often be explained by genetic factors. Findings imply similar etiologies of OA and several pain conditions.