Background There is evidence that knee pain not only is a consequence of structural deterioration in osteoarthritis (OA) but also contributes to structural progression. Clarifying this is important ...because targeting the factors related to knee pain may offer a clinical approach for slowing the progression of knee OA. The aim of this study was to examine whether knee pain over 1 year predicted cartilage volume loss, incidence and progression of radiographic osteoarthritis (ROA) over 4 years. Methods Osteoarthritis Initiative participants with no ROA (Kellgren-Lawrence grade less than or equai to 1) (n = 2120) and with ROA (Kellgren-Lawrence grade 2) (n = 2249) were examined. Knee pain was assessed at baseline and 1 year using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). Knee pain patterns were categorised as no pain (WOMAC pain < 5 at baseline and 1 year), fluctuating pain (WOMAC pain 5 at either time point) and persistent pain (WOMAC pain 5 at both time points). Cartilage volume, incidence and progression of ROA were assessed using magnetic resonance imaging and x-rays at baseline and 4-years. Results In both non-ROA and ROA, greater baseline WOMAC knee pain score was associated with increased medial and lateral cartilage volume loss (p less than or equai to 0.001), incidence (OR 1.07, 95% CI 1.01-1.13) and progression (OR 1.07, 95% CI 1.03-1.10) of ROA. Non-ROA and ROA participants with fluctuating and persistent knee pain had increased cartilage volume loss compared with those with no pain (p for trend less than or equai to 0.01). Non-ROA participants with fluctuating knee pain had increased risk of incident ROA (OR 1.62, 95% CI 1.04-2.54), corresponding to a number needed to harm of 19.5. In ROA the risk of progressive ROA increased in participants with persistent knee pain (OR 1.82, 95% CI 1.28-2.60), corresponding to a number needed to harm of 9.6. Conclusions Knee pain over 1 year predicted accelerated cartilage volume loss and increased risk of incident and progressive ROA. Early management of knee pain and controlling knee pain over time by targeting the underlying mechanisms may be important for preserving knee structure and reducing the burden of knee OA. Keywords: Pain, Knee osteoarthritis, Cartilage, Incidence, Progression, Magnetic resonance imaging
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.
Although intra-articular corticosteroid injections (IACI) are commonly used for the treatment of knee osteoarthritis (OA), there is controversy regarding possible deleterious effects on joint ...structure. In this line, this study investigates the effects of IACI on the evolution of knee OA structural changes and pain. Participants for this nested case-control study were from the Osteoarthritis Initiative. Knees of participants who had received an IACI and had magnetic resonance images (MRI) were named cases (n = 93), and each matched with one control (n = 93). Features assessed at the yearly visits and their changes within the follow-up period were from MRI (cartilage volume, meniscal thickness, bone marrow lesions, bone curvature, and synovial effusion size), X-ray (joint space width), and clinical (Western Ontario and McMaster Universities Osteoarthritis Index WOMAC pain score) data. Participants who received IACI experienced a transient and significantly greater rate of loss of the meniscal thickness (p = 0.006) and joint space width (p = 0.011) in the knee medial compartment in the year they received the injection, compared to controls. No significant effect of the IACI was found on the rate of cartilage loss nor on any other knee structural changes or WOMAC pain post-treatment. In conclusion, a single IACI in knee OA was shown to be safe with no negative impact on structural changes, but there was a transient meniscal thickness reduction, a phenomenon for which the clinical relevance is at present unknown.
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.
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.
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.
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.
There is an obvious need to identify biomarkers that could predict patient response to an osteoarthritis (OA) treatment. This post hoc study explored in a 2-year randomized controlled trial in ...patients with knee OA, the likelihood of some serum biomarkers to be associated with a better response to chondroitin sulfate in reducing cartilage volume loss.
Eight biomarkers were studied: hyaluronic acid (HA), C reactive protein (CRP), adipsin, leptin, N-terminal propeptide of collagen IIα (PIIANP), C-terminal crosslinked telopeptide of type I collagen (CTX-1), matrix metalloproteinase-1 (MMP-1), and MMP-3. Patients were treated with chondroitin sulfate (1200 mg/day; n = 57) or celecoxib (200 mg/day; n = 62). Serum biomarkers were measured at baseline. The cartilage volume at baseline and its loss at 2 years were assessed by quantitative magnetic resonance imaging (MRI). Statistical analysis included analysis of covariance.
As data from the original MOSAIC trial showed no differences in cartilage volume and loss in the lateral compartment of the knee joint between the two treatment groups in any comparison, only the medial compartment and its subregions were studied. Stratification according to the median biomarker levels was used to discriminate treatment effect. In patients with levels of biomarkers of inflammation (HA, leptin and adipsin) lower than the median, those treated with chondroitin sulfate demonstrated less cartilage volume loss in the medial compartment, condyle, and plateau (p ≤ 0.047). In contrast, patients treated with chondroitin sulfate with higher levels of MMP-1 and MMP-3, biomarkers of cartilage catabolism, had less cartilage volume loss in the medial compartment, condyle, and plateau (p ≤ 0.050). Patients with higher levels of PIIANP and CTX-1, biomarkers related to collagen anabolism and bone catabolism, respectively, had reduced cartilage volume loss in the medial condyle (p ≤ 0.026) in the chondroitin sulfate group.
This study is suggestive of a potentially greater response to chondroitin sulfate treatment on cartilage volume loss in patients with knee OA with low level of inflammation and/or greater level of cartilage catabolism.
This is a post hoc study. Original trial registration: ClinicalTrials.gov, NCT01354145 . Registered on 13 May 2011.
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.
Knee bone curvature assessed by MRI was associated with OA cartilage loss. A recent knee OA trial demonstrated the superiority of chondroitin sulfate over celecoxib (comparator) at reducing cartilage ...volume loss (CVL) in the medial compartment (condyle). The main objectives were to identify which baseline bone curvature regions of interest (BCROI) best associated with CVL and investigate whether baseline BCROI and 2-year change are correlated with the protective effect of chondroitin sulphate on CVL.
This post hoc analysis of a clinical trial used the according-to-protocol population (chondroitin sulphate, n = 57; celecoxib, n = 63) baseline and 2-year MRI to assess bone curvature and CVL. Global optimum search identified the BCROI in the medial condyle using celecoxib as reference. Statistical analyses were performed with Pearson's correlation, Mann-Whitney U -test, Student's t -test and analysis of covariance.
The BCROI including the medial posterior condyle and lateral central condyle was found to correlate best with medial condyle CVL at 2 years ( r = 0.33, P = 0.008). In patients with a baseline BCROI value less than the median (more flattened bone), chondroitin sulphate demonstrated a protective effect on CVL compared with celecoxib in the medial compartment (P = 0.037). In patients with 2-year BCROI changes greater than the median (greater severity of bone flattening), chondroitin sulphate protected against CVL in the medial compartment, condyle and central plateau (P ⩽ 0.030).
This study is the first to demonstrate the feasibility and usefulness of bone curvature measurements to predict effectiveness of OA treatment on CVL. The results identify bone curvature as a potential novel biomarker for knee OA clinical trials.
To examine whether metformin use was associated with knee cartilage volume loss over 4 years and risk of total knee replacement over 6 years in obese individuals with knee osteoarthritis.
This study ...analysed the Osteoarthritis Initiative participants with radiographic knee osteoarthritis (Kellgren-Lawrence grade ≥ 2) who were obese (body mass index BMI ≥ 30 kg/m
). Participants were classified as metformin users if they self-reported regular metformin use at baseline, 1-year and 2-year follow-up (n = 56). Non-users of metformin were defined as participants who did not report the use of metformin at any visit from baseline to 4-year follow-up (n = 762). Medial and lateral cartilage volume (femoral condyle and tibial plateau) were assessed using magnetic resonance imaging at baseline and 4 years. Total knee replacement over 6 years was assessed. General linear model and binary logistic regression were used for statistical analyses.
The rate of medial cartilage volume loss was lower in metformin users compared with non-users (0.71% vs. 1.57% per annum), with a difference of - 0.86% per annum (95% CI - 1.58% to - 0.15%, p = 0.02), after adjustment for age, gender, BMI, pain score, Kellgren-Lawrence grade, self-reported diabetes, and weight change over 4 years. Metformin use was associated with a trend towards a significant reduction in risk of total knee replacement over 6 years (odds ratio 0.30, 95% CI 0.07-1.30, p = 0.11), after adjustment for age, gender, BMI, Kellgren-Lawrence grade, pain score, and self-reported diabetes.
These data suggest that metformin use may have a beneficial effect on long-term knee joint outcomes in those with knee osteoarthritis and obesity. Randomised controlled trials are needed to confirm these findings and determine whether metformin would be a potential disease-modifying drug for knee osteoarthritis with the obese phenotype.
To explore, using MRI, the disease-modifying effect of strontium ranelate (SrRan) treatment on cartilage volume loss (CVL) and bone marrow lesions (BMLs) in a subset of patients from a Phase III ...clinical trial in knee osteoarthritis (OA) (SrRan Efficacy in Knee OsteoarthrItis triAl (SEKOIA)).
Patients with primary symptomatic knee OA were randomised to receive either SrRan 1 g/day or 2 g/day or placebo (SEKOIA study). A subset of these patients had MRIs at baseline, 12, 24 and 36 months to assess the knee cartilage volume and BMLs. Missing values were imputed and the analyses were adjusted according to Bonferroni.
In this MRI subset, the distribution of patients (modified intention-to-treat; n=330) was 113, 105 and 112 for SrRan 1 g/day, 2 g/day and placebo, respectively. The groups were fairly balanced at baseline regarding demographics, clinical symptoms or imaging characteristics. Treatment with SrRan 2 g/day significantly decreased CVL on the plateaus at 12 (p=0.002) and 36 (p=0.003) months compared with placebo. Of note, in the medial femur and plateau, SrRan 1 g/day, but not SrRan 2 g/day, had more CVL than placebo. In patients with BML in the medial compartment at baseline, the BML score at 36 months was decreased in both treatment groups compared with the placebo group (SrRan 1 g/day, p=0.002 and SrRan 2 g/day p=0.001, respectively), and CVL significantly decreased with SrRan 2 g/day (p=0.023) in the plateau compared with placebo.
In knee OA patients, treatment with SrRan 2 g/day was found to have beneficial effects on structural changes by significantly reducing CVL in the plateau and BML progression in the medial compartment.
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.