Radiographic Knee Osteoarthritis Research Articles

Artificial intelligence tools trained on human-labeled data reflect human biases: a case study in a large clinical consecutive knee osteoarthritis cohort

Humans have been shown to have biases when reading medical images, raising questions about whether humans are uniform in their disease gradings. Artificial intelligence (AI) tools trained on human-labeled data may have inherent human non-uniformity. In this study, we used a radiographic knee osteoarthritis external validation dataset of 50 patients and a six-year retrospective consecutive clinical cohort of 8,273 patients. An FDA-approved and CE-marked AI tool was tested for potential non-uniformity in Kellgren-Lawrence grades between the right and left sides of the images. We flipped the images horizontally so that a left knee looked like a right knee and vice versa. According to human review, the AI tool showed non-uniformity with 20–22% disagreements on the external validation dataset and 13.6% on the cohort. However, we found no evidence of a significant difference in the accuracy compared to senior radiologists on the external validation dataset, or age bias or sex bias on the cohort. AI non-uniformity can boost the evaluated performance against humans, but image areas with inferior performance should be investigated.

Quantitative analysis of effectiveness and associated factors of exercise on symptoms in osteoarthritis: a pharmacodynamic model-based meta-analysis

ObjectiveThis study aims to evaluate the time point and magnitude of peak effectiveness of exercise and the effects of various exercise modalities for osteoarthritis (OA) symptoms and to identify factors...

Noisy knees - knee crepitus prevalence and association with structural pathology: a systematic review and meta-analysis

ObjectiveKnee crepitus, the audible crackling or grinding noise during knee movement, can be experienced across the lifespan and create concern for underlying pathology. Our systematic review aims to provide a...

Influence of Knee Osteoarthritis Severity, Knee Pain, and Depression on Physical Function: A Cross-Sectional Study.

Decreased physical function with increasing life expectancy is a public health concern worldwide. Knee osteoarthritis (KOA) is considered one of the primary illnesses causing decreased physical function. Depression affects decreased physical function and is closely related to knee pain in KOA. However, the effect of these interacting factors on physical function is not clear. We conducted a cross-sectional analysis of the baseline data of 1106 subjects of the 2009 Locomotive Syndrome and Health Outcome in Aizu Cohort Study (LOHAS). We determined the association between their Timed Up and Go test (TUG) scores and radiographic KOA, knee pain, and depression in a multivariate analysis. Severe knee pain was significantly associated with decreased physical function (the odds ratio [OR] was 2.13, 95% confidence interval [CI]: 1.32-4.89), as was depression (OR 2.64, 95% CI 1.61-4.33). Only Kellgren-Lawrence (KL) grade 4 was significantly associated with decreased physical function in the radiographic KOA severity (OR 6.58, 95% CI 1.75-24.68). Severe knee pain and depression were significantly associated with decreased physical function, but not radiographic KOA severity except for KL grade 4. The limitations of using radiographic KOA severity alone as the indicator of assessment for physical function were suggested. When assessing decreased physical function, the clinical focus tends to be on radiographic KOA severity, but it is important to consider the patient's knee pain and psychological factors.

Development of the machine learning model that is highly validated and easily applicable to predict radiographic knee osteoarthritis progression.

Many models using the aid of artificial intelligence have been recently proposed to predict the progression of knee osteoarthritis. However, previous models have not been properly validated with an external data set or have reported poor predictive performances. Therefore, the purpose of this study was to design a machine learning model for knee osteoarthritis progression, focusing on high validation quality and clinical applicability. A retrospective analysis was conducted on prospectively collected data, using the Osteoarthritis Initiative data set (5966 knees) for model development and the Multicenter Osteoarthritis Study data set (3392 knees) for validation. The analysis aimed to predict Kellgren-Lawrence grade (KLG) progression over 4-5 years in knees with initial KLG of 0, 1, or 2. Possible predictors included demographics, comorbidities, history of meniscectomy, gait speed, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores, and radiological findings. The Random Forest algorithm was employed for the predictive model development. Baseline KLG, contralateral knee osteoarthritis, lateral joint space narrowing (JSN) grade, BMI, medial JSN grade, and total WOMAC score were six features selected for the model in descending order of importance. Odds ratios of baseline KLG, contralateral knee osteoarthritis, and lateral JSN grade were 1.76, 2.59, and 4.74, respectively (all p < 0.001). The area-under-the-curve of the ROC curve in the validation set was 0.76 with an accuracy of 0.68 and an F1-score of 0.56. The progression of knee osteoarthritis in 4 ~ 5 years could be well-predicted using easily available variables. This simple and validated model may aid surgeons in knee osteoarthritis patient management.

Do Existing MRI Definitions of Knee Osteoarthritis Identify Knees That Will Develop Clinically Significant Disease Over Up To 11 Years of Follow-Up?

In individuals without radiographic knee osteoarthritis (OA), we investigated whether magnetic resonance imaging (MRI)-defined knee OA at baseline was associated with incident radiographic and symptomatic disease during up to 11 years of follow-up. Osteoarthritis Initiative participants without tibiofemoral radiographic knee OA at baseline were assessed for MRI-based tibiofemoral cartilage damage, osteophyte presence, bone marrow lesions, and meniscal damage/extrusion. We defined MRI knee OA using alternative, reported definitions (Def A and Def B). Kellgren-Lawrence (KL) grade, joint space narrowing (JSN), and frequent knee symptoms (Sx) were assessed at baseline, 1-, 2-, 3-, 4-, 6-, 8-, and 10/11-year follow-up visits. Incident tibiofemoral radiographic knee OA (outcome) was defined as (1) KL ≥2, (2) KL ≥2 and JSN, or (3) KL ≥2 and Sx. Adjusted Cox proportional hazards regression models examined associations of baseline MRI-defined knee OA (Def A and Def B) with incident outcomes during up to 11 years of follow-up. Among 1,621 participants (mean age ± SD 58.8 ± 9.0 years, mean body mass index ± SD 27.2 ± 4.5 kg/m2, 59.5% women), 17% had MRI-defined knee OA by Def A and 24% by Def B. Baseline MRI-defined knee OA was associated with incident KL ≥2 (odds ratio 2.94 [95% confidence interval (95% CI) 2.34-3.68] for Def A and 2.44 [95% CI 1.97-3.03] for Def B). However, a substantial proportion of individuals with baseline MRI-defined knee OA did not develop incident KL ≥2 during follow-up (59% for Def A and 64% for Def B). Findings were similar for the other two outcomes. Current MRI definitions of knee OA do not adequately identify knees that will develop radiographic and symptomatic disease.

Association between folate intake and radiographic progression, pain function scores in subjects with radiographic knee osteoarthritis: Data from the osteoarthritis initiative.

Folate has an important role in the functioning of the musculoskeletal system, including modulation of inflammation, immunity, cartilage regeneration, prevention of osteoporosis, and maintenance of muscle strength, but evidence on the association between folate intake and knee pain, functional scores, and radiographic progression in patients with knee osteoarthritis (OA) is still limited. Our population-based cohort was extracted from the osteoarthritis initiative (OAI), focusing on individuals with prevalent radiographic knee OA (with a Kellgren-Lawrence score ≥2). Folate consumption was determined using the food frequency questionnaire. Data regarding the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores and radiographic readings were collected over 48 months. We analyzed the compiled data using generalized additive mixed models. Our cohort consisted of 1472 OA patients (626 men and 846 women, mean [SD] age 62.35 [8.92]). At the 48-month follow-up, we observed a significant correlation between higher folate intake and a slower progression of knee pain and functional scores, as evidenced by a statistically significant decrease in the WOMAC total score, WOMAC pain subscale score, and WOMAC function/disability subscale score (p < .05). The fully adjusted models estimated a reduction of -0.028 points per 50 μg/1000 kcal of daily folate intake on the WOMAC pain subscale, -0.117 points on the WOMAC function subscale, and -0.160 points on the total WOMAC scale. Furthermore, our nonparametric fit analysis suggested that a higher intake of folate might decelerate the radiographic progression of OA. Stratified analyses indicated that an increase in folate consumption might particularly benefit men, older adults, overweight and obese individuals, and those with a higher dietary fiber intake. Higher folate intake is correlated with improved knee function and reduced pain in patients with knee OA and might deter the radiographic progression of OA. The benefits appear to be more pronounced in men, older adults, overweight and obese individuals, and those with a higher dietary fiber intake.

Pain Reduction With Oral Methotrexate in Knee Osteoarthritis : A Randomized, Placebo-Controlled Clinical Trial.

Treatments for osteoarthritis (OA) are limited. Previous small studies suggest that the antirheumatic drug methotrexate may be a potential treatment for OA pain. To assess symptomatic benefits of methotrexate in knee OA (KOA). A multicenter, randomized, double-blind, placebo-controlled trial done between 13 June 2014 and 13 October 2017. (ISRCTN77854383; EudraCT: 2013-001689-41). 15 secondary care musculoskeletal clinics in the United Kingdom. A total of 207 participants with symptomatic, radiographic KOA and knee pain (severity ≥4 out of 10) on most days in the past 3 months with inadequate response to current medication were approached for inclusion. Participants were randomly assigned 1:1 to oral methotrexate once weekly (6-week escalation 10 to 25 mg) or matched placebo over 12 months and continued usual analgesia. The primary end point was average knee pain (numerical rating scale [NRS] 0 to 10) at 6 months, with 12-month follow-up to assess longer-term response. Secondary end points included knee stiffness and function outcomes and adverse events (AEs). A total of 155 participants (64% women; mean age, 60.9 years; 50% Kellgren-Lawrence grade 3 to 4) were randomly assigned to methotrexate (n = 77) or placebo (n = 78). Follow-up was 86% (n = 134; methotrexate: 66, placebo: 68) at 6 months. Mean knee pain decreased from 6.4 (SD, 1.80) at baseline to 5.1 (SD, 2.32) at 6 months in the methotrexate group and from 6.8 (SD, 1.62) to 6.2 (SD, 2.30) in the placebo group. The primary intention-to-treat analysis showed a statistically significant pain reduction of 0.79 NRS points in favor of methotrexate (95% CI, 0.08 to 1.51; P = 0.030). There were also statistically significant treatment group differences in favor of methotrexate at 6 months for Western Ontario and McMaster Universities Osteoarthritis Index stiffness (0.60 points [CI, 0.01 to 1.18]; P = 0.045) and function (5.01 points [CI, 1.29 to 8.74]; P = 0.008). Treatment adherence analysis supported a dose-response effect. Four unrelated serious AEs were reported (methotrexate: 2, placebo: 2). Not permitting oral methotrexate to be changed to subcutaneous delivery for intolerance. Oral methotrexate added to usual medications demonstrated statistically significant reduction in KOA pain, stiffness, and function at 6 months. Versus Arthritis.

Progression of Bone Marrow Lesions and the Development of Knee Osteoarthritis: Osteoarthritis Initiative Data.

Background Bone marrow lesions (BMLs) are a known risk factor for incident knee osteoarthritis (OA), and deep learning (DL) methods can assist in automated segmentation and risk prediction. Purpose To develop and validate a DL model for quantifying tibiofemoral BML volume on MRI scans in knees without radiographic OA and to assess the association between longitudinal BML changes and incident knee OA. Materials and Methods This retrospective study included knee MRI scans from the Osteoarthritis Initiative prospective cohort (February 2004-October 2015). The DL model, developed between August and October 2023, segmented the tibiofemoral joint into 10 subregions and measured BML volume in each subregion. Baseline and 4-year follow-up MRI scans were analyzed. Knees without OA at baseline were categorized into three groups based on 4-year BML volume changes: BML-free, BML regression, and BML progression. The risk of developing radiographic and symptomatic OA over 9 years was compared among these groups. Results Included were 3869 non-OA knees in 2430 participants (mean age, 59.5 years ± 9.0 [SD]; female-to-male ratio, 1.3:1). At 4-year follow-up, 2216 knees remained BML-free, 1106 showed an increase in BML volume, and 547 showed a decrease in BML volume. BML progression was associated with a higher risk of developing radiographic knee OA compared with remaining BML-free (hazard ratio [HR] = 3.0; P < .001) or BML regression (HR = 2.0; P < .001). Knees with BML progression also had a higher risk of developing symptomatic OA compared with BML-free knees (HR = 1.3; P < .001). Larger volume changes in BML progression were associated with a higher risk of developing both radiographic OA (HR = 2.0; P < .001) and symptomatic OA (HR = 1.7; P < .001). In almost all subchondral plates, especially the medial femur and tibia, BML progression was associated with a higher risk of developing both radiographic and symptomatic OA compared with remaining BML-free. Conclusion Knees with BML progression, according to subregion and extent of volume changes, were associated with an increased risk of OA compared with BML-free knees and knees with BML regression, highlighting the potential utility of monitoring BML volume changes in evaluating interventions to prevent OA development. ClinicalTrials.gov Identifier: NCT00080171 © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Said and Sakly in this issue.

A Review of Current Approaches to Pain Management in Knee Osteoarthritis with a Focus on Italian Clinical Landscape.

The global cases of knee osteoarthritis (KOA) are projected to increase by 74.9% by 2050. Currently, over half of patients remain dissatisfied with their pain relief. This review addresses unmet needs for moderate-to-severe KOA pain; it offers evidence and insights for improved management. Italian experts from the fields of rheumatology, physical medicine and rehabilitation, orthopedics, primary care, and pain therapy have identified several key issues. They emphasized the need for standardized care protocols to address inconsistencies in patient management across different specialties. Early diagnosis is crucial, as cartilage responds better to early protective and structural therapies. Faster access to physiatrist evaluation and reimbursement for physical, rehabilitative, and pharmacological treatments, including intra-articular (IA) therapy, could reduce access disparities. Concerns surround the adverse effects of oral pharmacological treatments, highlighting the need for safer alternatives. Patient satisfaction with corticosteroids and hyaluronic acid-based IA therapies reduces over time and there is no consensus on the optimal IA therapy protocol. Surgery should be reserved for severe symptoms and radiographic KOA evidence, as chronic pain post-surgery poses significant societal and economic burdens. The experts advocate for a multidisciplinary approach, promoting interaction and collaboration between specialists and general practitioners, to enhance KOA care and treatment consistency in Italy.

Radiographic Knee Osteoarthritis Is a Risk Factor for the Development of Dementia: Locomotive Syndrome and Health Outcomes in the Aizu Cohort Study.

Osteoarthritis is linked to dementia, but no longitudinal studies have established this connection. This prospective cohort study from the Locomotive Syndrome and Health Outcome in Aizu Cohort Study (LOHAS) aimed to determine if knee osteoarthritis (KOA) independently predicts dementia in adults aged 65 and above. Participants were classified by the Kellgren-Laurence scale into no/minimal KOA (grades 0 and I) and definitive KOA (grade II or higher). We analyzed dementia incidence from 2009 to 2015 using long-term care insurance data, adjusting for age, sex, vascular risks, depressive symptoms, and activity levels. Out of 1089 participants (58.9% female, average age 72.5), 72.0% had definitive KOA. Dementia occurrence was significantly higher in the definitive group (8.4%) compared to the no/minimal group (3.0%) (p < 0.001). A log-rank test and Cox regression analysis confirmed these findings, showing an adjusted hazard ratio of 2.29 (confidence interval: 1.12-4.68) for dementia in those with definitive KOA. These results suggest that KOA is a significant risk factor for dementia, highlighting the importance of addressing contributing factors in KOA patients to potentially slow the progression of dementia.

Knee muscle strength and steadiness for individuals with anterior cruciate ligament reconstruction and knee osteoarthritis

BackgroundStrength and neuromuscular decrements following knee musculoskeletal injury may accelerate knee osteoarthritis development. This study assessed isometric knee extensor and flexor strength and steadiness between individuals with knee injury, i.e., ligament reconstruction, and knee osteoarthritis to healthy age-matched controls. MethodsFour cohorts (1: knee injury and 2: age-matched controls, and 3: radiographic knee osteoarthritis and 4: age-matched controls) were recruited. Participants performed maximal voluntary isometric knee extensor and flexor contractions. Then, strength (e.g., peak and rate of torque development) and steadiness (e.g., peak power, mean, and median frequency) were derived from each raw torque-time curve and associated power spectral density. A Kruskal-Wallis H test and Spearman's rho correlation analysis assessed cohort differences and association between knee extensor and flexor strength and steadiness. FindingsThe young adult control and knee injury cohorts exhibited greater knee extensor and flexor strength than the older, knee osteoarthritis cohort (p < 0.043). The knee injury cohort, despite being as strong as their healthy counterparts, were significantly less steady with a 92% increase in peak power frequency (p = 0.046). The osteoarthritis cohort exhibited 157% less total power compared to the knee injury and young control cohorts (p < 0.019). Knee extensor and flexor peak torque, rate of torque development, and mean torque exhibit a significant, positive relation with total power (p < 0.018). InterpretationIndividuals with knee injury and disease may exhibit weaker or less steady knee musculature, predisposing them to degenerative joint disease. Clinicians may need to restore knee extensor and flexor steadiness to facilitate better joint neuromuscular control.

Accurate, automated classification of radiographic knee osteoarthritis severity using a novel method of deep learning: Plug-in modules

BackgroundFine-grained classification deals with data with a large degree of similarity, such as cat or bird species, and similarly, knee osteoarthritis severity classification [Kellgren–Lawrence (KL) grading] is one such fine-grained classification task. Recently, a plug-in module (PIM) that can be integrated into convolutional neural-network-based or transformer-based networks has been shown to provide strong discriminative regions for fine-grained classification, with results that outperformed the previous deep learning models. PIM utilizes each pixel of an image as an independent feature and can subsequently better classify images with minor differences. It was hypothesized that, as a fine-grained classification task, knee osteoarthritis severity may be classified well using PIMs. The aim of the study was to develop this automated knee osteoarthritis classification model.MethodsA deep learning model that classifies knee osteoarthritis severity of a radiograph was developed utilizing PIMs. A retrospective analysis on prospectively collected data was performed. The model was trained and developed using the Osteoarthritis Initiative dataset and was subsequently tested on an independent dataset, the Multicenter Osteoarthritis Study (test set size: 17,040). The final deep learning model was designed through an ensemble of four different PIMs.ResultsThe accuracy of the model was 84%, 43%, 70%, 81%, and 96% for KL grade 0, 1, 2, 3, and 4, respectively, with an overall accuracy of 75.7%.ConclusionsThe ensemble of PIMs could classify knee osteoarthritis severity using simple radiographs with a fine accuracy. Although improvements will be needed in the future, the model has been proven to have the potential to be clinically useful.

Predicting incident radiographic knee osteoarthritis through quantitative meniscal lesion parameters: data from the osteoarthritis initiative

BackgroundThis study investigates the potential of novel meniscal parameters as predictive factors for incident radiographic knee osteoarthritis (ROA) over a span of four years, as part of the Osteoarthritis Initiative (OAI) study.ObjectivesQuantitative measurements of meniscal parameters alteration could serve as predictors of OA’s occurrence and progression.Methods and materialsA nested matched case-control study design was used to select participants from OAI study. Case knees (n = 178) were defined as those with incident ROA (Kellgren Lawrence Grade (KLG) 0 or 1 at baseline (BL), evolving into KLG 2 or above by year 4). Control knees were matched one-to-one by sex, age and radiographic status with case knees. The mean distance from medial-to-lateral meniscal lesions [Mean(MLD)], mean value of tibial plateau width [Mean(TPW)] and the mean of the relative percentage of the medial-to-lateral meniscal lesions distance [Mean(RMLD)] were evaluated through coronal T2-weighted turbo spin echo (TSE) MRI at P-0 (visit when incident ROA was found on radiograph), P-1(one year prior to P-0) and baseline, respectively. Using the imaging data of one patient, the mechanism was investigated by finite element analysis.ResultsParticipants were on average 60.22 years old, predominantly female (66.7%) and overweight (mean BMI: 28.15). Mean(MLD) and Mean(RMLD) were significantly greater for incident knees compared to no incident knees at baseline, P-1 and P-0. [Mean(MLD), Mean(RMLD); (42.56–49.73) mean ± (7.70–9.52) mm SD vs. (38.14–40.78) mean ± (5.51–7.05)mm SD; (58.61–68.95) mean ± (8.52–11.40) mm SD vs. (52.52–56.35) mean ± (6.53–7.85)mm SD, respectively]. Baseline Mean(MLD) and Mean(RMLD), [Adjusted OR, 95%CI: 1.11(1.07 to 1.16) and 1.13(1.09 to 1.17), respectively], were associated with incident ROA during 4 years, However, Mean(TPW) [Adjusted OR, 95%CI: 0.98(0.94 to 1.02)] was not associated with incident ROA during 4 years. While Mean(TPW) at P-1 and P-0 was not associated with the risk of incident ROA, Mean(MLD) and Mean(RMLD) at P-1 and P-0 were significantly positively associated with the risk of incident ROA.ConclusionsThe meniscal parameters alteration could be an important imaging biomarker to predict the occurrence of ROA.

Association of serum biomarkers with radiographic knee osteoarthritis, knee pain and function in a young, male, trauma-exposed population – Findings from the ADVANCE study

ObjectiveThe ArmeD SerVices TrAuma RehabilitatioN OutComE (ADVANCE) study is investigating long-term combat-injury outcomes; this sub-study aims to understand the association of osteoarthritis (OA) biomarkers with knee radiographic OA (rOA), pain and function in this high-risk population for post-traumatic OA. DesignADVANCE compares combat-injured participants with age, rank, deployment and job-role frequency-matched uninjured participants. Post-injury immunoassay-measured serum biomarkers, knee radiographs, knee injury and osteoarthritis outcome scale (KOOS), and six-minute walk-tests (6MWT) are reported. The primary analysis, adjusted for age, body mass, socioeconomic status, and ethnicity, was to determine any differences in biomarkers between those with/without combat-injury, rOA and pain. Secondary analyses were performed to compare post-traumatic/idiopathic OA, painful/painfree rOA and injury patterns. Results1145 male participants were recruited, aged 34.1±5.4, 8.9±2.2 years post-injury (n=579 trauma-exposed, of which, traumatic-amputation n=161) or deployment (n=566 matched). Cartilage oligomeric matrix protein (COMP) was significantly higher in the combat-injured group compared to uninjured (p=0.01). Notably, COMP was significantly lower in the traumatic-amputation group compared to non-amputees (p<0.001), decreasing relative to number of amputations (p<0.001). Leptin was higher (p=0.005) and adiponectin lower (p=0.017) in those with v without knee pain, associated with an increased risk of 22% and 17% for pain, and 46% and 34% for painful rOA, respectively. There were no significant differences between trauma-exposed and unexposed participants with rOA. ConclusionsThe most notable findings of this large, unique study are the similarities between those with rOA regardless of trauma-exposure, the injury-pattern and traumatic-amputation-associated differences in COMP, and the relationship between adipokines and pain.

Infrapatellar fat pad size and subcutaneous fat in knee osteoarthritis radiographic progression: data from the osteoarthritis initiative

ObjectivesAdipose tissue has been associated with knee osteoarthritis (KOA) pathogenesis, but the longitudinal changes in adipose tissue with KOA progression have not been carefully evaluated. This study aimed to determine if longitudinal changes of systemic and local adipose tissue is associated with radiographic progression of KOA.MethodsThis case-control study used data from the Osteoarthritis Initiative (OAI) and included 315 cases (all the right knees with a minimum of Kellgren-Lawrence score (KL) of 0 and an increase of ≥ 1 KL from baseline to 48 months) and 315 controls matched by age, sex, race, and baseline KL. Cross sectional area of IPFP (IPFP CSA) and subcutaneous adipose tissue around the distal thigh (SCATthigh) were measured using MRI images at baseline and 24 months. Conditional logistic regression models were fitted to estimate associations of obesity markers, IPFP CSA, and SCATthigh with radiographic KOA progression. Mediation analysis was used to assess whether IPFP CSA or SCATthigh mediates the relationships between baseline BMI and radiographic KOA progression.Results24-month changes of IPFP CSA (ΔIPFP CSA) and SCATthigh (ΔSCATthigh) were significantly greater in cases compared to controls, whereas Δ BMI and Δ abdominal circumference were similar in both groups during follow-up. Adjusted ORs for radiographic KOA progression were 9.299, 95% CI (5.357–16.141) per 1 SD increase of Δ IPFP CSA and 1.646, 95% CI (1.288–2.103) per 1 SD increase of Δ SCATthigh. ΔIPFP CSA mediated the association between baseline BMI and radiographic KOA progression (87%).ConclusionsSubjects with radiographic progression of KOA, had significant increases in IPFP CSA and subcutaneous adipose tissue while BMI and abdominal circumference remained stable. Additional studies are needed to confirm these associations.

Integrating Radiomics and Neural Networks for Knee Osteoarthritis Incidence Prediction.

Accurately predicting knee osteoarthritis (KOA) is essential for early detection and personalized treatment. We aimed to develop and test a magnetic resonance imaging (MRI)-based joint space (JS) radiomic model (RM) to predict radiographic KOA incidence through neural networks by integrating meniscus and femorotibial cartilage radiomic features. In the Osteoarthritis Initiative cohort, participants with knees without radiographic KOA at baseline but at high risk for radiographic KOA were included. Patients' knees developed radiographic KOA, whereas control knees did not over four years. We randomly split the participants into development and test cohorts (8:2) and extracted features from baseline three-dimensional double-echo steady-state sequence MRI. Model performance was evaluated using an area under the receiver operating characteristic curve (AUC), sensitivity, and specificity in both cohorts. Nine resident surgeons performed the reader experiment without/with the JS-RM aid. Our study included 549 knees in the development cohort (275 knees of patients with KOA vs 274 knees of controls) and 137 knees in the test cohort (68 knees of patients with KOA vs 69 knees of controls). In the test cohort, JS-RM had a favorable accuracy for predicting the radiographic KOA incidence with an AUC of 0.931 (95% confidence interval [CI] 0.876-0.963), a sensitivity of 84.4% (95% CI 83.9%-84.9%), and a specificity of 85.6% (95% CI 85.2%-86.0%). The mean specificity and sensitivity of resident surgeons through MRI reading in predicting radiographic KOA incidence were increased from 0.474 (95% CI 0.333-0.614) and 0.586 (95% CI 0.429-0.743) without the assistance of JS-RM to 0.874 (95% CI 0.847-0.901) and 0.812 (95% CI 0.742-0.881) with JS-RM assistance, respectively (P < 0.001). JS-RM integrating the features of the meniscus and cartilage showed improved predictive values in radiographic KOA incidence.

Associations between static foot posture, dynamic in-shoe plantar foot forces and knee pain in people with medial knee osteoarthritis: A cross-sectional exploratory study

ObjectiveTo investigate relationships between static foot posture, dynamic plantar foot forces and knee pain in people with medial knee osteoarthritis (OA). DesignData from 164 participants with symptomatic, moderate to severe radiographic medial knee OA were analysed. Knee pain was self-reported using a numerical rating scale (NRS; scores 0-10; higher scores worse) and the Knee Injury and Osteoarthritis Outcome Score pain subscale (KOOS; scores 0-100; lower scores worse). Static foot posture was assessed using clinical tests (foot posture index, foot mobility magnitude, navicular drop). Dynamic plantar foot forces (lateral, medial, whole foot, medial-lateral ratio, arch index) were measured using an in-shoe plantar pressure system while walking. Relationships between foot posture and plantar forces (independent variables) and pain (dependent variables) were evaluated using linear regression models, unadjusted and adjusted for sex, walking speed, KL grade, shoe category, and body mass (for dynamic plantar foot forces). ResultsNo measure of static foot posture was associated with any knee pain measure. Higher medial-lateral foot force ratio at midstance, and a higher arch index during overall stance, were weakly associated with higher knee pain on the NRS (regression coefficient=0.69, 95% confidence interval (CI) 0.09 to 1.28) and KOOS (coefficient=3.03, 95% CI 0.71 to 5.35) pain scales, respectively. ConclusionDynamic plantar foot forces, but not static foot posture, were associated with knee pain in people with medial knee OA. However, the amount of pain explained by increases in plantar foot force was small, thus these associations are unlikely to be clinically meaningful.

Interobserver Agreement and Performance of Concurrent AI Assistance for Radiographic Evaluation of Knee Osteoarthritis.

Background Due to conflicting findings in the literature, there are concerns about a lack of objectivity in grading knee osteoarthritis (KOA) on radiographs. Purpose To examine how artificial intelligence (AI) assistance affects the performance and interobserver agreement of radiologists and orthopedists of various experience levels when evaluating KOA on radiographs according to the established Kellgren-Lawrence (KL) grading system. Materials and Methods In this retrospective observer performance study, consecutive standing knee radiographs from patients with suspected KOA were collected from three participating European centers between April 2019 and May 2022. Each center recruited four readers across radiology and orthopedic surgery at in-training and board-certified experience levels. KL grading (KL-0 = no KOA, KL-4 = severe KOA) on the frontal view was assessed by readers with and without assistance from a commercial AI tool. The majority vote of three musculoskeletal radiology consultants established the reference standard. The ordinal receiver operating characteristic method was used to estimate grading performance. Light kappa was used to estimate interrater agreement, and bootstrapped t statistics were used to compare groups. Results Seventy-five studies were included from each center, totaling 225 studies (mean patient age, 55 years ± 15 [SD]; 113 female patients). The KL grades were KL-0, 24.0% (n = 54); KL-1, 28.0% (n = 63); KL-2, 21.8% (n = 49); KL-3, 18.7% (n = 42); and KL-4, 7.6% (n = 17). Eleven readers completed their readings. Three of the six junior readers showed higher KL grading performance with versus without AI assistance (area under the receiver operating characteristic curve, 0.81 ± 0.017 [SEM] vs 0.88 ± 0.011 [P < .001]; 0.76 ± 0.018 vs 0.86 ± 0.013 [P < .001]; and 0.89 ± 0.011 vs 0.91 ± 0.009 [P = .008]). Interobserver agreement for KL grading among all readers was higher with versus without AI assistance (κ = 0.77 ± 0.018 [SEM] vs 0.85 ± 0.013; P < .001). Board-certified radiologists achieved almost perfect agreement for KL grading when assisted by AI (κ = 0.90 ± 0.01), which was higher than that achieved by the reference readers independently (κ = 0.84 ± 0.017; P = .01). Conclusion AI assistance increased junior readers' radiographic KOA grading performance and increased interobserver agreement for osteoarthritis grading across all readers and experience levels. Published under a CC BY 4.0 license. Supplemental material is available for this article.

Prolonged quadriceps latency during gait early after anterior cruciate ligament injury predicts radiographic knee osteoarthritis 6-years after anterior cruciate ligament reconstruction

BackgroundThe purpose was to explore quadriceps electromechanical function (quadriceps latency) during gait after anterior cruciate ligament injury as a predictor for radiographic knee osteoarthritis 6-years after anterior cruciate ligament reconstruction. Change in latency after preoperative physical therapy was also examined. MethodsQuadriceps latency (time between peak knee moment and quadriceps electromyography) was calculated before preoperative physical therapy (2.4 [0.5–7.5] months after anterior cruciate ligament injury) and after preoperative physical therapy in 24 athletes. Participants were dichotomized into osteoarthritis (Kellgren and Lawrence grade ≥ 2) and non-osteoarthritis groups at 6-years. Forward selection logistic regression was performed using z-score normalized quadriceps latency and demographics. A 2 × 2 repeated measure ANOVA was performed for quadriceps latency between groups before and after preoperative physical therapy. FindingsQuadriceps latency before preoperative physical therapy was the only predictor of 6-year radiographic osteoarthritis (p = 0.014, odds ratio [95% confidence interval] = 5.859 [1.435–23.924]). Time by group interaction was observed for quadriceps latency (p = 0.039, η2p = 0.179). In the osteoarthritis group, latency may reduce after training (before preoperative physical therapy = 115.7 ± 20.6 ms, after preoperative physical therapy = 99.5 ± 24.0 ms, p = 0.082). InterpretationProlonged latency after anterior cruciate ligament injury may predict post-traumatic knee osteoarthritis 6-years after anterior cruciate ligament reconstruction. Latency may shorten with preoperative physical therapy, yet athletes still moved on to develop osteoarthritis. Quadriceps function may need intervention immediately following anterior cruciate ligament injury for prevention of post-traumatic knee osteoarthritis.