Prediction models for the risk of total knee replacement: development and validation using data from multicentre cohort studies

Radiomics analysis of patellofemoral joint improves knee replacement risk prediction: Data from the Multicenter Osteoarthritis Study (MOST)

Objective

Knee replacement (KR) is the last-resort treatment for knee osteoarthritis. Although radiographic evidence of tibiofemoral joint has been widely adopted for prognostication, patellofemoral joint has gained little attention and may hold additional value for further improvements. We aimed to quantitatively analyse patellofemoral joint through radiomics analysis of lateral view radiographs for improved KR risk prediction.

Design

From the Multicenter Osteoarthritis Study dataset, we retrospectively retrieved the initial-visit lateral left knee radiographs of 2943 patients aged 50 to 79. They were split into training and test cohorts at a 2:1 ratio. A comprehensive set of radiomic features were extracted within the best-performing subregion of patellofemoral joint and combined into a radiomics score (RadScore). A KR risk score, derived from Kellgren-Lawrence grade (KLG) of tibiofemoral joint and RadScore of patellofemoral joint, was developed by multivariate Cox regression and assessed using time-dependent area under receiver operating characteristic curve (AUC).

Results

While patellofemoral osteoarthritis (PFOA) was insignificant during multivariate analysis, RadScore was identified as an independent risk factor (multivariate Cox p-value < 0.001) for KR. The subgroup analysis revealed that RadScore was particularly effective in predicting rapid progressor (KR occurrence before 30 months) among early- (KLG < 2) and mid-stage (KLG ​= ​2) patients. Combining two joints radiographic information, the AUC reached 0.89/0.87 for predicting 60-month KR occurrence.

Conclusions

The RadScore of the patellofemoral joint on lateral radiographs emerges as an independent prognostic factor for improving KR prognosis prediction. The KR risk score could be instrumental in managing progressive knee osteoarthritis interventions.

An interpretable knee replacement risk assessment system for osteoarthritis patients

Objective

Knee osteoarthritis (OA) is a complex disease with heterogeneous representations. Although it is modifiable to prevention and early treatment, there still lacks a reliable and accurate prognostic tool. Hence, we aim to develop a quantitative and self-administrable knee replacement (KR) risk stratification system for knee osteoarthritis (KOA) patients with clinical features.

Method

A total of 14 baseline features were extracted from 9592 cases in the Osteoarthritis Initiative (OAI) cohort. A survival model was constructed using the Random Survival Forests algorithm. The prediction performance was evaluated with the concordance index (C-index) and average receiver operating characteristic curve (AUC). A three-class KR risk stratification system was built to differentiate three distinct KR-free survival groups. Thereafter, Shapley Additive Explanations (SHAP) was introduced for model explanation.

Results

KR incidence was accurately predicted by the model with a C-index of 0.770 (±0.0215) and an average AUC of 0.807 (±0.0181) with 14 clinical features. Three distinct survival groups were observed from the ten-point KR risk stratification system with a four-year KR rate of 0.79%, 5.78%, and 16.2% from the low, medium, and high-risk groups respectively. KR is mainly caused by pain medication use, age, surgery history, diabetes, and a high body mass index, as revealed by SHAP.

Conclusion

A self-administrable and interpretable KR survival model was developed, underscoring a KR risk scoring system to stratify KOA patients. It will encourage regular self-assessments within the community and facilitate personalised healthcare for both primary and secondary prevention of KOA.

Incidence of Total Knee Arthroplasty After Arthroscopic Surgery for Knee Osteoarthritis: A Secondary Analysis of a Randomized Clinical Trial

Importance

It is unclear whether arthroscopic resection of degenerative knee tissues among patients with osteoarthritis (OA) of the knee delays or hastens total knee arthroplasty (TKA); opposite findings have been reported.

Objective

To compare the long-term incidence of TKA in patients with OA of the knee after nonoperative management with or without additional arthroscopic surgery.

Design, Setting, and Participants

In this ad hoc secondary analysis of a single-center, assessor-blinded randomized clinical trial performed from January 1, 1999, to August 31, 2007, 178 patients were followed up through March 31, 2019. Participants included adults diagnosed with OA of the knee referred for potential arthroscopic surgery in a tertiary care center specializing in orthopedics in London, Ontario, Canada. All participants from the original randomized clinical trial were included. Data were analyzed from June 1, 2021, to October 20, 2022.

Exposures

Arthroscopic surgery (resection or debridement of degenerative tears of the menisci, fragments of articular cartilage, or chondral flaps and osteophytes that prevented full extension) plus nonoperative management (physical therapy plus medications as required) compared with nonoperative management only (control).

Main Outcomes and Measures

Total knee arthroplasty was identified by linking the randomized trial data with prospectively collected Canadian health administrative datasets where participants were followed up for a maximum of 20 years. Multivariable Cox proportional hazards regression models were used to compare the incidence of TKA between intervention groups.

Results

A total of 178 of 277 eligible patients (64.3%; 112 [62.9%] female; mean [SD] age, 59.0 [10.0] years) were included. The mean (SD) body mass index was 31.0 (6.5). With a median follow-up of 13.8 (IQR, 8.4-16.8) years, 31 of 92 patients (33.7%) in the arthroscopic surgery group vs 36 of 86 (41.9%) in the control group underwent TKA (adjusted hazard ratio [HR], 0.85 [95% CI, 0.52-1.40]). Results were similar when accounting for crossovers to arthroscopic surgery (13 of 86 [15.1%]) during follow-up (HR, 0.88 [95% CI, 0.53-1.44]). Within 5 years, the cumulative incidence was 10.2% vs 9.3% in the arthroscopic surgery group and control group, respectively (time-stratified HR for 0-5 years, 1.06 [95% CI, 0.41-2.75]); within 10 years, the cumulative incidence was 23.3% vs 21.4%, respectively (time-stratified HR for 5-10 years, 1.06 [95% CI, 0.45-2.51]). Sensitivity analyses yielded consistent results.

Conclusions and Relevance

In this secondary analysis of a randomized clinical trial of arthroscopic surgery for patients with OA of the knee, a statistically significant association with delaying or hastening TKA was not identified. Approximately 80% of patients did not undergo TKA within 10 years of nonoperative management with or without additional knee arthroscopic surgery.

Trial Registration

ClinicalTrials.gov Identifier: NCT00158431.

Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative

The prevalence of knee osteoarthritis (OA) is widespread and the heterogeneous patient factors and clinical symptoms in OA patients impede developing personalized treatments for OA patients. In this study, we used unsupervised and supervised machine learning to organize the heterogeneity in knee OA patients and predict disease progression in individuals from the Osteoarthritis Initiative (OAI) dataset. We identified four distinct knee OA phenotypes using unsupervised learning that were defined by nutrition, disability, stiffness, and pain (knee and back) and were strongly related to disease fate. Interestingly, the absence of supplemental vitamins from an individual's diet was protective from disease progression. Moreover, we established a phenotyping tool and prognostic model from 5 variables (WOMAC disability score of the right knee, WOMAC total score of the right knee, WOMAC total score of the left knee, supplemental vitamins and minerals frequency, and antioxidant combination multivitamins frequency) that can be utilized in clinical practice to determine the risk of knee OA progression in individual patients. We also developed a prognostic model to estimate the risk for total knee replacement and provide suggestions for modifiable variables to improve long-term knee health. This combination of unsupervised and supervised data-driven tools provides a framework to identify knee OA phenotype in a clinical scenario and personalize treatment strategies.

Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative

The prevalence of knee osteoarthritis (OA) is widespread and the heterogeneous patient factors and clinical symptoms in OA patients impede developing personalized treatments for OA patients. In this study, we used unsupervised and supervised machine learning to organize the heterogeneity in knee OA patients and predict disease progression in individuals from the Osteoarthritis Initiative (OAI) dataset. We identified four distinct knee OA phenotypes using unsupervised learning that were defined by nutrition, disability, stiffness, and pain (knee and back) and were strongly related to disease fate. Interestingly, the absence of supplemental vitamins from an individual's diet was protective from disease progression. Moreover, we established a phenotyping tool and prognostic model from 5 variables (WOMAC disability score of the right knee, WOMAC total score of the right knee, WOMAC total score of the left knee, supplemental vitamins and minerals frequency, and antioxidant combination multivitamins frequency) that can be utilized in clinical practice to determine the risk of knee OA progression in individual patients. We also developed a prognostic model to estimate the risk for total knee replacement and provide suggestions for modifiable variables to improve long-term knee health. This combination of unsupervised and supervised data-driven tools provides a framework to identify knee OA phenotype in a clinical scenario and personalize treatment strategies.

The Prognostic Potential of End-Stage Knee Osteoarthritis and Its Components to Predict Knee Replacement: Data From the Osteoarthritis Initiative

OBJECTIVE

We aimed to determine how 2 definitions of end-stage knee osteoarthritis (esKOA) and each component (knee symptoms, persistent knee pain, radiographic severity, and presence of limited mobility or instability) related to future knee replacement (KR).

METHODS

We performed knee-based analyses of Osteoarthritis Initiative data from baseline to the first 4 annual follow-up visits, and data on KR from baseline until the fifth yearly contact. We calculated a base model using common risk factors for KR in logistic regression models with generalized estimating equations. We assessed model performance with area under the receiver-operating characteristic curve (AUC) and Hosmer-Lemeshow test. We then added esKOA or each component from the visit (< 12 months) before a KR and change in the year before a KR. We calculated the net reclassification improvement (NRI) index and the integrated discrimination improvement (IDI) index.

RESULTS

Our sample was mostly female (58%), ≥ 65 years old, White (82%), and without radiographic knee osteoarthritis (50%). At the visit before a KR, Kellgren-Lawrence (KL) grades (ordinal scale; AUC 0.88, NRI 1.12, IDI 0.11), the alternate definition of esKOA (AUC 0.84, NRI 1.16, IDI 0.12), and a model with every component of esKOA (AUC 0.91, NRI 1.30, IDI 0.17) had the best performances. During the year before a KR, change in esKOA status (alternate definition) had the best performance (AUC 0.86, NRI 1.24, IDI 0.12).

CONCLUSION

Radiographic severity may be a screening tool to find a knee that will likely receive a KR. However, esKOA may be an ideal outcome in clinical trials because a change in esKOA state predicts future KR.

Charting Aging Trajectories of Knee Cartilage Thickness for Early Osteoarthritis Risk Prediction: An MRI Study from the Osteoarthritis Initiative Cohort

Knee osteoarthritis (OA), a prevalent joint disease in the U.S., poses challenges in terms of predicting of its early progression. Although high-resolution knee magnetic resonance imaging (MRI) facilitates more precise OA diagnosis, the heterogeneous and multifactorial aspects of OA pathology remain significant obstacles for prognosis. MRI-based scoring systems, while standardizing OA assessment, are both time-consuming and labor-intensive. Current AI technologies facilitate knee OA risk scoring and progression prediction, but these often focus on the symptomatic phase of OA, bypassing initial-stage OA prediction. Moreover, their reliance on complex algorithms can hinder clinical interpretation. To this end, we make this effort to construct a computationally efficient, easily-interpretable, and state-of-the-art approach aiding in the radiographic OA (rOA) auto-classification and prediction of the incidence and progression, by contrasting an individual's cartilage thickness with a similar demographic in the rOA-free cohort. To better visualize, we have developed the toolset for both prediction and local visualization. A movie demonstrating different subtypes of dynamic changes in local centile scores during rOA progression is available at https://tli3.github.io/KneeOA/. Specifically, we constructed age-BMI-dependent reference charts for knee OA cartilage thickness, based on MRI scans from 957 radiographic OA (rOA)-free individuals from the Osteoarthritis Initiative cohort. Then we extracted local and global centiles by contrasting an individual's cartilage thickness to the rOA-free cohort with a similar age and BMI. Using traditional boosting approaches with our centile-based features, we obtain rOA classification of KLG ≤ 1 versus KLG = 2 (AUC = 0.95, F1 = 0.89), KLG ≤ 1 versus KLG ≥ 2 (AUC = 0.90, F1 = 0.82) and prediction of KLG2 progression (AUC = 0.98, F1 = 0.94), rOA incidence (KLG increasing from < 2 to ≥ 2; AUC = 0.81, F1 = 0.69) and rOA initial transition (KLG from 0 to 1; AUC = 0.64, F1 = 0.65) within a future 48-month period. Such performance in classifying KLG ≥ 2 matches that of deep learning methods in recent literature. Furthermore, its clinical interpretation suggests that cartilage changes, such as thickening in lateral femoral and anterior femoral regions and thinning in lateral tibial regions, may serve as indicators for prediction of rOA incidence and early progression. Meanwhile, cartilage thickening in the posterior medial and posterior lateral femoral regions, coupled with a reduction in the central medial femoral region, may signify initial phases of rOA transition.

Improved 2-Year Freedom from Arthroplasty in Patients with High-Risk SIFK Scores and Medial Knee Osteoarthritis Treated with an Implantable Shock Absorber versus Non-Operative Care

OBJECTIVE

Subchondral insufficiency fracture of the knee (SIFK) is associated with high rates of osteoarthritis (OA) and arthroplasty. The implantable shock absorber (ISA) is an extra-capsular implant that unloads the medial knee compartment. This study compared the 2-year freedom from arthroplasty rates in subjects with medial knee OA and SIFK when treated with an ISA versus a matched cohort of patients treated non-surgically.

DESIGN

This retrospective case-control study compared 2-year conversion rates to arthroplasty in SIFK score-, age-, and body mass index (BMI)-matched control subjects without prior surgical history with ISA-implanted subjects from an ongoing prospective study. Baseline and final radiographs, and MRIs were reviewed for evaluation of meniscus or ligament injuries, insufficiency fractures, and subchondral edema. Kaplan-Meier analysis assessed survival.

RESULTS

Forty-two patients (21 Control: 21 ISA), mean age = 52.3 ± 8.7 years, BMI = 29.5 ± 3.9 kg/m2, 40% female were evaluated. Both ISA and Control arms had the same numbers of low (n = 4), medium (n = 11), and high-risk (n = 6) SIFK scores. One- and 2-year freedom-from-arthroplasty rates were both 100% for ISA subjects, and 76% and 55%, respectively, for Controls (P = 0.001 for cross-group comparison). Control knees with low, medium, and high-risk SIFK scores had respective 1- and 2-year survival rates of 100% and 100%, 90% and 68% (P = 0.07 vs. ISA), and 33% and 0% (P = 0.002 vs. ISA).

CONCLUSIONS

ISA intervention was strongly associated with avoidance of arthroplasty at a minimum 2 years, especially in patients with high-risk SIFK scores. SIFK severity scoring predicted relative risk of conversion to arthroplasty through at least 2 years in non-surgically treated subjects.

An Interpretable Machine Learning Model for Predicting 10-Year Total Hip Arthroplasty Risk

BACKGROUND

As the demand for total hip arthroplasty (THA) rises, a predictive model for THA risk may aid patients and clinicians in augmenting shared decision-making. We aimed to develop and validate a model predicting THA within 10 years in patients using demographic, clinical, and deep learning (DL)-automated radiographic measurements.

METHODS

Patients enrolled in the Osteoarthritis Initiative were included. DL algorithms measuring osteoarthritis- and dysplasia-relevant parameters on baseline pelvis radiographs were developed. Demographic, clinical, and radiographic measurement variables were then used to train generalized additive models to predict THA within 10 years from baseline. A total of 4,796 patients were included (9,592 hips; 58% female; 230 THAs (2.4%)). Model performance using 1) baseline demographic and clinical variables 2) radiographic variables, and 3) all variables were compared.

RESULTS

Using 110 demographic and clinical variables, the model had a baseline area under the receiver operating curve (AUROC) of 0.68 and area under the precision recall curve (AUPRC) of 0.08. Using 26 DL-automated hip measurements, the AUROC was 0.77 and AUPRC was 0.22. Combining all variables, the model improved to an AUROC of 0.81 and AUPRC of 0.28. Three of the top five predictive features in the combined model were radiographic variables including minimum joint space along with hip pain and analgesic use. Partial dependency plots revealed predictive discontinuities for radiographic measurements consistent with literature thresholds of osteoarthritis progression and hip dysplasia.

CONCLUSION

A machine learning model predicting 10-year THA performed more accurately with DL radiographic measurements. The model weighted predictive variables in concordance with clinical THA-pathology assessments.

Willingness to Have Total Knee Arthroplasty in Rural Areas of Northern China

Objective

To evaluate willingness and its potential predictors to have total knee arthroplasty (TKA) among residents in rural areas of northern China.

Methods

Data were collected from two population‐based studies on osteoarthritis (OA) in northern China. Residents aged ≥ 50 years in randomly selected rural communities were recruited using a cluster survey method. Participants completed a home interview (including social‐demographic characteristics, clinical information, 12‐Item Short Form Health Surveys [SF‐12], and Visual Analog Scale [VAS] of knee pain), a physical examination, and bilateral weight‐bearing posteroanterior semi‐flexed view of radiographs of knees. Willingness to have TKA was queried by asking: “If you need to undergo total knee arthroplasty as judged by orthopaedic surgeons, are you willing to accept this operation?” Two trained investigators read all radiographs independently and reached an acceptable intra‐reader agreement. We fitted univariate and multivariate regressions adjusting for potential confounders to examine predictors of willingness to have TKA stratified by sex.

Results

A total of 2172 participants were included. The overall rate of willingness to have TKA was 33.8%. Men were more likely to be willing to have TKA than women with an odds ratio (OR) of 0.73 (95% confidence interval [CI]: 0.60–0.89, P = 0.002). A higher household income (OR: 2.34 for men and 1.77 for women, both P < 0.001), higher SF‐12 Physical Component Summary (PCS) score (OR: 1.02 for both gender, P = 0.04 for men and P = 0.01 for women), and being aware of TKA (OR: 2.55 for men and 2.65 for women, both P < 0.001) were associated with a higher willingness to have TKA. Other predictors of higher willingness to have TKA were younger age (OR: 0.66 for 60–70 years, P = 0.01; and 0.48 for >70 years, P = 0.003, respectively) and lower level of physical activity (OR: 0.57 for moderate, p = 0.04; and 0.62 for heavy, p = 0.04, respectively) among men and a higher education (OR: 1.45 for Junior school, P = 0.04; and 1.66 for high school and above, P = 0.02, respectively) and being overweight among women (OR: 1.38, P = 0.008), respectively. No significant difference was observed between willingness to have TKA and frequent knee pain, VAS of knee pain and Kellgren and Lawrence grades in both men and women.

Conclusions

The willingness to have TKA among rural residents of northern China was relatively low. Younger age, women, educational level, household income, physical function, and awareness of TKA were positively associated with willingness to have TKA.