Development and validation of prediction models to estimate risk of primary total hip and knee replacements using data from the UK: two prospective open cohorts using the UK Clinical Practice Research Datalink

Accuracy of machine learning in identifying candidates for total knee arthroplasty (TKA) surgery: a systematic review and meta-analysis

BACKGROUND

The application of machine learning (ML) in predicting the requirement for total knee arthroplasty (TKA) at knee osteoarthritis (KOA) patients has been acknowledged. Nonetheless, the variables employed in the development of ML models are diverse and these different approaches yield inconsistent predictive performance of models. Therefore, we conducted this systematic review and meta-analysis to explore the feasibility of ML in identifying candidates for TKA.

METHOD

This study was conducted based on the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. This study was registered on the international prospective register of systematic reviews registration database website, PROSPERO, with a unique ID: CRD 42023443948. The study subjects were patients diagnosed with KOA. Relevant studies were searched through PubMed, Web of Science, Cochrane, and Embase until September 15, 2024. The c-index was used as the outcome measure. The risk of bias in the primary study was assessed by Prediction model Risk of Bias Assessment Tool (PROBAST). Random or fixed effects were used for the meta-analysis.

RESULTS

A total of 13 articles were included in this study, but only 11 articles with 25 models were eligible for the meta-analysis. ML models in the included studies were classified based on the source of variables, including clinical features, radiomics, and the combination of clinical features and radiomics. In the training set, the c-index was 0.713 (0.628 - 0.799) for clinical features, 0.841 (0.777 - 0.904) for radiomics, and 0.844 (0.815 - 0.873) for the combination of clinical features and radiomics. In the validation set, the c-index for ML models based on clinical features, radiomics, and the combination of clinical features and radiomics was 0.656 (0.526 - 0.786), 0.861 (0.806 - 0.916), and 0.831 (0.799 - 0.863), respectively.

CONCLUSION

The results of this meta-analysis highlighted that the ML model is feasible in identifying candidates for TKA. X-ray-based ML models exhibit the best predictive performance among the models. However, there is currently a lack of high-level research available for clinical application. Furthermore, the accuracy of ML models in identifying candidates for TKA is significantly limited by the quality of modeling parameters and database architecture. Therefore, constructing a more targeted and professional database is imperative to promote the development and clinical application of ML models.

Metabolic profiling of smoking, associations with type 2 diabetes and interaction with genetic susceptibility

BACKGROUND

Smokers are at increased risk of type 2 diabetes (T2D), but the underlying mechanisms are unclear. We investigated if the smoking-T2D association is mediated by alterations in the metabolome and assessed potential interaction with genetic susceptibility to diabetes or insulin resistance.

METHODS

In UK Biobank (n = 93,722), cross-sectional analyses identified 208 metabolites associated with smoking, of which 131 were confirmed in Mendelian Randomization analyses, including glycoprotein acetyls, fatty acids, and lipids. Elastic net regression was applied to create a smoking-related metabolic signature. We estimated hazard ratios (HR) of incident T2D in relation to baseline smoking/metabolic signature and calculated the proportion of the smoking-T2D association mediated by the signature. Additive interaction between the signature and genetic risk scores for T2D (GRS-T2D) and insulin resistance (GRS-IR) on incidence of T2D was assessed as relative excess risk due to interaction (RERI).

FINDINGS

The HR of T2D was 1·73 (95% confidence interval (CI) 1·54 - 1·94) for current versus never smoking, and 38·3% of the excess risk was mediated by the metabolic signature. The metabolic signature and its mediation role were replicated in TwinGene. The metabolic signature was associated with T2D (HR: 1·61, CI 1·46 - 1·77 for values above vs. below median), with evidence of interaction with GRS-T2D (RERI: 0·81, CI: 0·23 - 1·38) and GRS-IR (RERI 0·47, CI: 0·02 - 0·92).

INTERPRETATION

The increased risk of T2D in smokers may be mediated through effects on the metabolome, and the influence of such metabolic alterations on diabetes risk may be amplified in individuals with genetic susceptibility to T2D or insulin resistance.

Development and Validation of a Deep-Learning Model to Predict Total Hip Replacement on Radiographs: The Total Hip Replacement Prediction (THREP) Model

BACKGROUND

There are few methods for accurately assessing the risk of total hip arthroplasty (THA) in patients with osteoarthritis. A novel and reliable method that could play a substantial role in research and clinical routine should be investigated. The purpose of the present study was to develop a deep-learning model that can reliably predict the risk of THA with use of radiographic images and clinical symptom data.

METHODS

This retrospective, multicenter, case-control study assessed hip joints on weighted-bearing anteroposterior pelvic radiographs obtained from Osteoarthritis Initiative (OAI) participants. Participants who underwent THA were matched to controls according to age, sex, body mass index, and ethnicity. Cases and controls were uniformly split into training, validation, and testing data sets at proportions of 72% (n = 528), 14% (n = 104), and 14% (n = 104), respectively. Images and clinical symptom data were passed through a detection model and a deep convolutional neural network (DCNN) model to predict the probability of THA within 9 years as well as the most likely time period for THA (0 to 2 years, 3 to 5 years, 6 to 9 years). Model performance was assessed with use of the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity in the testing set.

RESULTS

A total of 736 participants were evaluated, including 184 cases and 552 controls. The prediction model achieved an overall accuracy, sensitivity, and specificity of 91.35%, 92.59% and 86.96%, respectively, with an AUC of 0.944, for THA within 9 years. The AUC of the DCNN model for assessing the most likely time period was 0.907 for 0 to 2 years, 0.916 for 3 to 5 years, and 0.841 for 6 to 9 years. Gradient-weighted class activation mapping closely corresponded to regions affecting the prediction of the DCNN model.

CONCLUSIONS

The proposed DCNN model is a reliable and valid method to predict the probability of THA-within limitations. It could assist clinicians in patient counseling and decision-making regarding the timing of the intervention. In the future, by increasing the size of the data set, enhancing the ethnic and socioeconomic diversity of the participants, and improving the follow-up rate, the quality of the conclusions can be further improved.

LEVEL OF EVIDENCE

Prognostic Level III . See Instructions for Authors for a complete description of levels of evidence.

A scoping review and best evidence synthesis for treatment of partial-thickness rotator cuff tears

BACKGROUND

Rotator cuff disorders include a broad spectrum of pathological conditions including partial-thickness and full-thickness tears. Studies have shown partial-thickness rotator cuff tear (PTRCT) prevalence to be twice that of full-thickness tears. In the working population, PTRCTs are one of the most common causes of shoulder pain and often result in occupational disability due to pain, stiffness, and loss of shoulder function. Treatment of PTRCTs remains controversial. The purpose of this study was to consolidate the existing high-quality evidence on best management approaches in treating PTRCTs using both nonoperative and operative approaches.

METHODS

A scoping review with best evidence synthesis was performed as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews. MEDLINE (OVID), EMBASE (OVID), Cochrane Library (Wiley), SCOPUS, Web of Science Core Collection, CINAHL Plus with Full Text (EBSCOhost), PubMed Central, and Science Direct were searched from 2000 to March 3, 2023. Level 1 studies, and systematic reviews and meta-analyses that included level 1 and 2 studies, were included.

RESULTS

The search yielded 8276 articles. A total of 3930 articles were screened after removing 4346 duplicates. Application of inclusion criteria resulted in 662 articles that were selected for full-text review. Twenty-eight level 1 studies, 1 systematic review, 4 meta-analyses, and 1 network meta-analyses were included in the best evidence synthesis. Nonoperative strategies included injections (ie, platelet-rich plasma, corticosteroid, prolotherapy, sodium hyaluronate, anesthetic, and atelocollagen), exercise therapy, and physical agents. Operative interventions consisted of débridement, shaving of the tendon and footprint, transtendon repair, and traditional suture anchor repair techniques with and without tear completion. Both nonoperative and operative strategies demonstrated effectiveness at managing pain and functional outcome for PTRCTs. The evidence supports the effectiveness of surgical intervention in treating PTRCTs regardless of arthroscopic technique.

CONCLUSION

The results of this scoping review do not support superiority of operative over nonoperative management and suggest that both strategies can be effective at managing pain and functional outcome for PTRCTs. Surgery, however, is the most invasive and costly approach, with the highest risk of complications such as infection. Other variables such as patient expectation, treating practitioner bias, or preference may change which modalities are offered and in what sequence.

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.

Real-world evidence that among atrial fibrillation patients warfarin is associated with reduced nonelective admissions compared with those on DOACs

BACKGROUND

Randomized trials show inconsistent estimates on risks of direct-acting oral anticoagulants (DOACs) versus warfarin in bleeding and mortality for atrial fibrillation (AF) patients. Trials are confounded by additional DOAC adherence support, while warfarin has a low time in therapeutic range. Few real-world studies compared emergency hospitalization risk between DOAC and warfarin users in AF. This study aimed to determine emergency hospitalization risk for AF patients on DOACs or warfarin in real-world settings.

METHODS

A tapered-matched real-world cohort extracted data from 412 English general practices' primary care records linked with emergency department (ED) and hospitalization data from the ECLIPSE database. AF patients with new DOAC or warfarin prescriptions were included. The primary outcome was all-cause ED attendance; the secondary outcomes were ED re-attendance, nonelective hospitalization, and rehospitalization within 12 months. Weighted Cox regression estimated relative risk difference.

RESULTS

39 201 DOAC patients were matched with 13 145 warfarin patients. DOAC patients had a 25% higher likelihood of attending ED (odds ratio 1.25; 95% confidence interval [CI] 1.01-1.55). DOAC use also associated with higher ED re-attendance, nonelective hospitalization, and rehospitalization within 12 months: 1.41 (95% CI 1.00-1.98), 1.26 (1.00-1.57), and 1.54 (1.01-2.34), respectively, with p-values < .05.

CONCLUSIONS

DOACs for AF thromboprophylaxis are associated with the increased risk of ED attendance, recurrent hospitalization, and numerical rise in ED re-attendance and first nonelective hospitalization compared to warfarin. However, these real-world data cannot establish if this difference results from medication adherence, lack of regular DOAC clinic monitoring, unmeasured confounders, or fundamental agent efficacy disparities.

A machine learning approach reveals features related to clinicians' diagnosis of clinically relevant knee osteoarthritis

OBJECTIVES

To identify highly ranked features related to clinicians' diagnosis of clinically relevant knee OA.

METHODS

General practitioners (GPs) and secondary care physicians (SPs) were recruited to evaluate 5-10 years follow-up clinical and radiographic data of knees from the CHECK cohort for the presence of clinically relevant OA. GPs and SPs were gathered in pairs; each pair consisted of one GP and one SP, and the paired clinicians independently evaluated the same subset of knees. A diagnosis was made for each knee by the GP and SP before and after viewing radiographic data. Nested 5-fold cross-validation enhanced random forest models were built to identify the top 10 features related to the diagnosis.

RESULTS

Seventeen clinician pairs evaluated 1106 knees with 139 clinical and 36 radiographic features. GPs diagnosed clinically relevant OA in 42% and 43% knees, before and after viewing radiographic data, respectively. SPs diagnosed in 43% and 51% knees, respectively. Models containing top 10 features had good performance for explaining clinicians' diagnosis with area under the curve ranging from 0.76-0.83. Before viewing radiographic data, quantitative symptomatic features (i.e. WOMAC scores) were the most important ones related to the diagnosis of both GPs and SPs; after viewing radiographic data, radiographic features appeared in the top lists for both, but seemed to be more important for SPs than GPs.

CONCLUSIONS

Random forest models presented good performance in explaining clinicians' diagnosis, which helped to reveal typical features of patients recognized as clinically relevant knee OA by clinicians from two different care settings.

Predicting total knee replacement at 2 and 5 years in osteoarthritis patients using machine learning

Objectives

Knee osteoarthritis is a major cause of physical disability and reduced quality of life, with end-stage disease often treated by total knee replacement (TKR). We set out to develop and externally validate a machine learning model capable of predicting the need for a TKR in 2 and 5 years time using routinely collected health data.

Design

A prospective study using datasets Osteoarthritis Initiative (OAI) and the Multicentre Osteoarthritis Study (MOST). OAI data were used to train the models while MOST data formed the external test set. The data were preprocessed using feature selection to curate 45 candidate features including demographics, medical history, imaging assessments, history of intervention and outcome.

Setting

The study was conducted using two multicentre USA-based datasets of participants with or at high risk of knee OA.

Participants

The study excluded participants with at least one existing TKR. OAI dataset included participants aged 45-79 years of which 3234 were used for training and 809 for internal testing, while MOST involved participants aged 50-79 and 2248 were used for external testing.

Main outcome measures

The primary outcome of this study was prediction of TKR onset at 2 and 5 years. Performance was evaluated using area under the curve (AUC) and F1-score and key predictors identified.

Results

For the best performing model (gradient boosting machine), the AUC at 2 years was 0.913 (95% CI 0.876 to 0.951), and at 5 years 0.873 (95% CI 0.839 to 0.907). Radiographic-derived features, questionnaire-based assessments alongside the patient's educational attainment were key predictors for these models.

Conclusions

Our approach suggests that routinely collected patient data are sufficient to drive a predictive model with a clinically acceptable level of accuracy (AUC>0.7) and is the first such tool to be externally validated. This level of accuracy is higher than previously published models utilising MRI data, which is not routinely collected.

The relationship between meniscal pathologies, cartilage loss, joint replacement and pain in knee osteoarthritis: a systematic review

OBJECTIVE

We conducted a systematic review in order to understand the relationship between imaging-visualised meniscus pathologies, hyaline cartilage, joint replacement and pain in knee osteoarthritis (OA).

DESIGN

A search of the Medline, Excerpta Medica database (EMBASE) and Cochrane library databases was performed for original publications reporting association between imaging-detected meniscal pathology (extrusion or tear/damage) and longitudinal and cross-sectional assessments of hyaline articular cartilage loss [assessed on magnetic resonance imaging (MRI)], incident joint replacement and pain (longitudinal and cross-sectional) in knee OA. Each association was qualitatively characterised by a synthesis of data from each analysis, based upon study design and quality scoring (including risk of bias assessment and adequacy of covariate adjustment using Cochrane recommended methodology).

RESULTS

In total 4,878 abstracts were screened and 82 publications were included (comprising 72 longitudinal analyses and 49 cross-sectional). Using high quality, well-adjusted data, meniscal extrusion and meniscal tear/damage were associated with longitudinal progression of cartilage loss, cross-sectional cartilage loss severity and joint replacement, independently of age, sex and body mass index (BMI). Medial and lateral meniscal tears were associated with cartilage loss when they occurred in the body and posterior horns, but not the anterior horns. There was a lack of high quality, well-adjusted meniscal pathology and pain publications and no clear independent association between meniscal extrusion or tear/damage with pain severity, progression in pain or incident frequent knee symptoms.

CONCLUSION

Meniscal features have strong associations with cartilage loss and joint replacement in knee OA, but weak associations with knee pain. Systematic review PROSPERO registration number: CRD 42020210910.

Proton pump inhibitor therapy and risk of knee replacement surgery: a general population-based cohort study

OBJECTIVE

Proton pump inhibitors (PPIs) are among the most commonly used medications for patients with osteoarthritis (OA). Various types of PPIs have different impacts on lowering serum magnesium level that may affect knee OA progression. We aimed to compare the risk of clinically relevant endpoint of knee replacement (KR) among initiators of five different PPIs with that among histamine-2 receptor antagonist (H2RA) initiators.

DESIGN

Among patients with knee OA (≥50 years) in The Health Improvement Network database in the UK we conducted five sequential propensity-score matched cohort studies to compare the risk of KR over 5-year among patients who initiated omeprazole (n = 2,672), pantoprazole (n = 664), lansoprazole (n = 3,747), rabeprazole (n = 751), or esomeprazole (n = 827) with those who initiated H2RA.

RESULTS

The prevalence of PPI prescriptions among participants with knee OA increased from 12.7% in 2000-44.0% in 2017. Two-hundred-and-seventy-four KRs (30.8/1,000 person-years) occurred in omeprazole initiators and 230 KRs (25.4/1,000 person-years) in H2RA initiators. Compared with H2RA initiators, the risk of KR was 21% higher in omeprazole initiators (hazard ratio [HR] = 1.21,95% confidence interval [CI]:1.01-1.44). Similar results were observed when pantoprazole use was compared with H2RA use (HR = 1.38,95%CI:1.00-1.90). No such an increased risk of KR was observed among lansoprazole (HR = 1.06,95%CI:0.92-1.23), rabeprazole (HR = 0.97,95%CI:0.73-1.30), or esomeprazole (HR = 0.83,95%CI:0.60-1.15) initiators compared with that among H2RA initiators.

CONCLUSIONS

In this population-based cohort study, initiation of omeprazole or pantoprazole use was associated with a higher risk of KR than initiation of H2RA use. This study raises concern regarding an unexpected risk of omeprazole and pantoprazole on accelerating OA progression.

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

Background

Few prognostic prediction models for total knee replacement are available, and the role of radiographic findings in predicting its use remains unclear. We aimed to develop and validate predictive models for total knee replacement and to assess whether adding radiographic findings improves predictive performance.

Methods

We identified participants with recent knee pain (in the past 3 months) in the Multicenter Osteoarthritis Study (MOST) and the Osteoarthritis Initiative (OAI). The baseline visits of MOST were initiated in 2003 and of OAI were initiated in 2004. We developed two predictive models for the risk of total knee replacement within 60 months of follow-up by fitting Cox proportional hazard models among participants in MOST. The first model included sociodemographic and anthropometric factors, medical history, and clinical measures (referred to as the clinical model). The second model added radiographic findings into the predictive model (the radiographic model). We evaluated each model's discrimination and calibration performance and assessed the incremental value of radiographic findings using both category-free net reclassification improvement (NRI) and integrated discrimination improvement (IDI). We tuned the models and externally validated them among participants in OAI.

Findings

We included 2658 participants from MOST (mean age 62·4 years [SD 8·1], 1646 [61·9%] women) in the training dataset and 4060 participants from OAI (mean age 60·9 years [9·1], 2379 [58·6%] women) in the validation dataset. 290 (10·9%) participants in the training dataset and 174 (4·3%) in the validation dataset had total knee replacement. The retained predictive variables included in the clinical model were age, sex, race, history of knee arthroscopy, frequent knee pain, current use of analgesics, current use of glucosamine, body-mass index, and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain score, and the most predictive factors were age, race, and WOMAC pain score. The retained predictive variables in the radiographic model were age, sex, race, frequent knee pain, current use of analgesics, WOMAC pain score, and Kellgren-Lawrence grade, and the most predictive factors were Kellgren-Lawrence grade, race, and age. The C-statistic was 0·79 (95% CI 0·76-0·81) for the clinical model and 0·87 (0·85-0·99) for the radiographic model in the training dataset. The calibration slope was 0·95 (95% CI 0·86-1·05) and 0·96 (0·87-1·04), respectively. Adding radiograph findings significantly improved predictive performance with an NRI of 0·43 (95% CI 0·38-0·50) and IDI of 0·14 (95% CI: 0·10-0·18). Both models, with tuned coefficients, showed a good predictive performance among participants in the validation dataset.

Interpretation

The risk of total knee replacement can be predicted based on common risk factors with good discrimination and calibration. Additionally, adding radiographic findings of knee osteoarthritis into the model substantially improves its predictive performance.

Funding

National Natural Science Foundation of China, National Key Research and Development Program, and Beijing Municipal Science & Technology Commission.

Digoxin targets low density lipoprotein receptor-related protein 4 and protects against osteoarthritis

OBJECTIVES

Dysregulated chondrocyte metabolism is closely associated with the pathogenesis of osteoarthritis (OA). Suppressing chondrocyte catabolism to restore cartilage homeostasis has been extensively explored, whereas far less effort has been invested toward enhancing chondrocyte anabolism. This study aimed to repurpose clinically approved drugs as potential stimulators of chondrocyte anabolism in treating OA.

METHODS

Screening of a Food and Drug Administration-approved drug library; Assays for examining the chondroprotective effects of digoxin in vitro; Assays for defining the therapeutic effects of digoxin using a surgically-induced OA model; A propensity-score matched cohort study using The Health Improvement Network to examine the relationship between digoxin use and the risk of joint OA-associated replacement among patients with atrial fibrillation; identification and characterisation of the binding of digoxin to low-density lipoprotein receptor-related protein 4 (LRP4); various assays, including use of CRISPR-Cas9 genome editing to delete LRP4 in human chondrocytes, for examining the dependence on LRP4 of digoxin regulation of chondrocytes.

RESULTS

Serial screenings led to the identification of ouabain and digoxin as stimulators of chondrocyte differentiation and anabolism. Ouabain and digoxin protected against OA and relieved OA-associated pain. The cohort study of 56 794 patients revealed that digoxin use was associated with reduced risk of OA-associated joint replacement. LRP4 was isolated as a novel target of digoxin, and deletion of LRP4 abolished digoxin's regulations of chondrocytes.

CONCLUSIONS

These findings not only provide new insights into the understanding of digoxin's chondroprotective action and underlying mechanisms, but also present new evidence for repurposing digoxin for OA.

Prognostic models for knee osteoarthritis: a protocol for systematic review, critical appraisal, and meta-analysis

BACKGROUND

Osteoarthritis is the most common degenerative joint disease. It is associated with significant socioeconomic burden and poor quality of life, mainly due to knee osteoarthritis (KOA), and related total knee arthroplasty (TKA). Since early detection method and disease-modifying drug is lacking, the key of KOA treatment is shifting to disease prevention and progression slowing. The prognostic prediction models are called for to guide clinical decision-making. The aim of our review is to identify and characterize reported multivariable prognostic models for KOA about three clinical concerns: (1) the risk of developing KOA in the general population, (2) the risk of receiving TKA in KOA patients, and (3) the outcome of TKA in KOA patients who plan to receive TKA.

METHODS

The electronic datasets (PubMed, Embase, the Cochrane Library, Web of Science, Scopus, SportDiscus, and CINAHL) and gray literature sources (OpenGrey, British Library Inside, ProQuest Dissertations & Theses Global, and BIOSIS preview) will be searched from their inception onwards. Title and abstract screening and full-text review will be accomplished by two independent reviewers. The multivariable prognostic models that concern on (1) the risk of developing KOA in the general population, (2) the risk of receiving TKA in KOA patients, and (3) the outcome of TKA in KOA patients who plan to receive TKA will be included. Data extraction instrument and critical appraisal instrument will be developed before formal assessment and will be modified during a training phase in advance. Study reporting transparency, methodological quality, and risk of bias will be assessed according to the TRIPOD statement, CHARMS checklist, and PROBAST tool, respectively. Prognostic prediction models will be summarized qualitatively. Quantitative metrics on the predictive performance of these models will be synthesized with meta-analyses if appropriate.

DISCUSSION

Our systematic review will collate evidence from prognostic prediction models that can be used through the whole process of KOA. The review may identify models which are capable of allowing personalized preventative and therapeutic interventions to be precisely targeted at those individuals who are at the highest risk. To accomplish the prediction models to cross the translational gaps between an exploratory research method and a valued addition to precision medicine workflows, research recommendations relating to model development, validation, or impact assessment will be made.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO CRD42020203543.

Development of a multivariable prediction model for early revision of total knee arthroplasty - The effect of including patient-reported outcome measures

BACKGROUND

Revision TKA is a serious adverse event with substantial consequences for the patient. As revision is becoming increasingly common in patients under 65 years, the need for improved preoperative patient selection is imminently needed. Therefore, this study aimed to identify the most important factors of early revision and to develop a prediction model of early revision including assessment of the effect of incorporating data on patient-reported outcome measures (PROMs).

MATERIAL AND METHODS

A cohort of 538 patients undergoing primary TKA was included. Multiple logistic regression using forward selection of variables was applied to identify the best predictors of early revision and to develop a prediction model. The model was internally validated with stratified 5-fold cross-validation. This procedure was repeated without including data on PROMs to develop a model for comparison. The models were evaluated on their discriminative capacity using area under the receiver operating characteristic curve (AUC).

RESULTS

The most important factors of early revision were age (OR 0.63 [0.42, 0.95]; P = 0.03), preoperative EQ-5D (OR 0.07 [0.01, 0.51]; P = 0.01), and number of comorbidities (OR 1.01 [0.97, 1.25]; P = 0.15). The AUCs of the models with and without PROMs were 0.65 and 0.61, respectively. The difference between the AUCs was not statistically significant (P = 0.32).

CONCLUSIONS

Although more work is needed in order to reach a clinically meaningful quality of the predictions, our results show that the inclusion of PROMs seems to improve the quality of the prediction model.

Effect of femoral posterior condyle offset on knee joint function after total knee replacement: a network meta-analysis and a sequential retrospective cohort study

BACKGROUND

This study was conducted with the aim to compare the effect of posterior condyle offset (PCO) changes on knee joint function of patients following total knee replacement (TKR).

METHODS

Electronic and manual searches were performed in the PubMed, Embase, and Cochrane Library databases from inception to September 2019. Network meta-analysis combined direct and indirect evidence to assess the weighted mean difference (WMD) and surface under the cumulative ranking curves (SUCRA) of different PCO changes (PCO ≤ - 2 mm, - 2 mm < PCO < 0 mm, 0 mm ≤ PCO < 2 mm and PCO ≥ 2 mm) on knee joint function after TKR. Then 103 OA patients undergoing unilateral TKR were included and the effect of PCO on the postoperative knee function was examined.

RESULTS

Totally, 5 cohort studies meeting the inclusion criteria were enrolled in this analysis. The results of meta-analysis showed that patients with 0 mm ≤ PCO < 2 mm after TKR had a better recovery of joint function (flexion contracture: 28.67%; KS functional score: 78.67%; KS knee score: 75.00%) than the remaining three groups. However, the knee flexion (77.00%) of patients with PCO ≤ - 2 mm after TKR was superior to the other three groups. Retrospective study also revealed a significant correlation between PCO changes and the flexion contracture, further flexion and KS functional score of patients after TKR, in which each functional knee score of patients with 0 mm ≤ PCO < 2 mm was better than the others.

CONCLUSION

These findings suggest a close correlation between PCO magnitude and knee joint function after TKR and that 0 mm ≤ PCO < 2 mm is superior to other changes for joint function after TKR.

Cardiovascular risks and bleeding with non-vitamin K antagonist oral anticoagulant versus warfarin in patients with type 2 diabetes: a tapered matching cohort study

BACKGROUND

We compared the risk of bleeding and cardiovascular disease (CVD) events between non-vitamin K antagonist oral anticoagulant (NOAC) and warfarin in people with type 2 diabetes (T2DM).

METHODS

862 Incident NOAC users and 626 incident warfarin users with T2DM were identified from within 40 UK general practice (1/4/2017-30/9/2018). Outcomes included incident hospitalisation for bleeding, CVD and re-hospitalisation for CVD within 12 months since first anticoagulant prescription, identified from linked hospitalisation data. A tapered matching method was applied to form comparison cohorts: coarsened exact matching restricted the comparison to areas of sufficient overlap in missingness and characteristics: (i) demographic characteristics; (ii) clinical measurements; (iii) prior bleeding and CVD history; (iv) prescriptions with bleeding; (v) anti-hypertensive treatment(s); (vi) anti-diabetes treatment(s). Entropy balancing sequentially balanced NOAC and warfarin users on their distribution of (i-vi). Weighted logistic regression modelling estimated outcome odds ratios (ORs), using entropy balancing weights from steps i-vi.

RESULTS

The 12-month ORs of bleeding with NOAC (n = 582) vs matched/balanced warfarin (n = 486) were 1.93 (95% confidence interval 0.97-3.84), 2.14 (1.03-4.44), 2.31 (1.10-4.85), 2.42 (1.14-5.14), 2.41 (1.12-5.18), and 2.51 (1.17-5.38) through steps i-vi. ORs for CVD re-hospitalisation was increased with NOAC treatment through steps i-vi: 2.21 (1.04-4.68), 2.13 (1.01-4.52), 2.47 (1.08-5.62), 2.46 (1.02-5.94), 2.51 (1.01-6.20), and 2.66 (1.02-6.94).

CONCLUSIONS

Incident NOAC use among T2DM is associated with increased risk of bleeding hospitalisation and CVD re-hospitalisation compared with incident warfarin use. For T2DM, caution is required in prescribing NOACs as first anticoagulant treatment. Further large-scale replication studies in external datasets are warranted.

Derivation and validation of a risk score predicting risk of early-onset peritonitis among patients initializing peritoneal dialysis: A cohort study

OBJECTIVES

Early onset peritonitis (EOP) increases the risk of clinical complications in patients initializing peritoneal dialysis (PD). This study aimed to develop and validate a risk prediction model for EOP among patients initializing PD.

METHODS

3772 patients registered with the Henan Peritoneal Dialysis Registry (HPDR) between 2007 and 2015 were included. The main outcome, EOP, was defined as incident peritonitis occurring within 6 months of the initialization of PD. Multivariable logistic regression modeling was applied to derive the risk score. All accessible clinical measurements were screened as potential predictors. Assessment of the developed model in terms of model discrimination and calibration was performed using C statistics and a calibration slope, respectively, and validated internally through a bootstrapping (1000-fold) method to adjust for over-fitting.

RESULTS

The absolute risk of EOP was 14.5%. Age, cardiac function measurements, serum electrolyte test items, lipid profiles, liver function test items, blood urea nitrogen, and white cell count were significant predictors of EOP in the final risk score. Good model discrimination, with C statistics above 0.70, and calibration of agreed observed and predicted risks were identified in the model.

CONCLUSION

A prediction model that quantifies risks of EOP has been developed and validated. It is based on a small number of clinical metabolic measurements that are available for patients initializing PD in many developing countries, and could serve as a tool to screen the population at high risk of EOP.

A Decline in Walking Speed Is Associated With Incident Knee Replacement in Adults With and at Risk for Knee Osteoarthritis

OBJECTIVE

To determine if a 1-year change in walking speed is associated with receiving an incident knee replacement during the following year in adults with and at risk for knee osteoarthritis (OA).

METHODS

Using data from the Osteoarthritis Initiative, we determined a 1-year change in the 20-meter walk speed from 3 observation periods (i.e., 0-12, 12-24, and 24-36 months). We operationally defined 1-year change in walking speed as either (1) decline: ≤ -0.1 m/s change, (2) no change: between -0.1 and 0.1 m/s change, and (3) increase: ≥ 0.1 m/s change. Incident knee replacement was defined using each subsequent 1-year period (i.e., 12-24, 24-36, and 36-48 months). Combining data from the 3 observation periods, we performed a Poisson regression with robust error variance to determine the relative risk between a change in walking speed (exposure) and incident knee replacement over the following year (outcome).

RESULTS

Of the 4264 participants included within this analysis (11,311 total person visits), 115 (3%) adults received a knee replacement. Decline in walking speed was associated with a 104% increase in risk [adjusted relative risk (RR) 2.04, 95% CI 1.40-2.98], while an increase in walking speed associated with a 55% decrease in risk (RR 0.45; 95% CI 0.22-0.93) of incident knee replacement in the following year compared to a person with no change in walking speed.

CONCLUSION

A 1-year decline in walking speed is associated with an increased risk, while a 1-year increase in walking speed is associated with a decreased risk of future incident knee replacement.

Predicting Total Knee Replacement from Symptomology and Radiographic Structural Change Using Artificial Neural Networks-Data from the Osteoarthritis Initiative (OAI)

The aim of the study was to longitudinally investigate symptomatic and structural factors prior to total knee replacement (TKR) surgery in order to identify influential factors that can predict a patient's need for TKR surgery. In total, 165 participants (60% females; 64.5 ± 8.4 years; 29.7 ± 4.7 kg/m²) receiving a TKR in any of both knees within a four-year period were analyzed. Radiographic change, knee pain, knee function and quality of life were annually assessed prior to the TKR procedure. Self-learning artificial neural networks were applied to identify driving factors for the surgical procedure. Significant worsening of radiographic structural change was observed prior to TKR (p ≤ 0.0046), whereas knee symptoms (pain, function, quality of life) worsened significantly only in the year prior to the TKR procedure. By using our prediction model, we were able to predict correctly 80% of the classified individuals to undergo TKR surgery with a positive predictive value of 84% and a negative predictive value of 73%. Our prediction model offers the opportunity to assess a patient's need for TKR surgery two years in advance based on easily available patient data and could therefore be used in a primary care setting.

Deep Learning Predicts Total Knee Replacement from Magnetic Resonance Images

Knee Osteoarthritis (OA) is a common musculoskeletal disorder in the United States. When diagnosed at early stages, lifestyle interventions such as exercise and weight loss can slow OA progression, but at later stages, only an invasive option is available: total knee replacement (TKR). Though a generally successful procedure, only 2/3 of patients who undergo the procedure report their knees feeling “normal” post-operation, and complications can arise that require revision. This necessitates a model to identify a population at higher risk of TKR, particularly at less advanced stages of OA, such that appropriate treatments can be implemented that slow OA progression and delay TKR. Here, we present a deep learning pipeline that leverages MRI images and clinical and demographic information to predict TKR with AUC 0.834 ± 0.036 (p < 0.05). Most notably, the pipeline predicts TKR with AUC 0.943 ± 0.057 (p < 0.05) for patients without OA. Furthermore, we develop occlusion maps for case-control pairs in test data and compare regions used by the model in both, thereby identifying TKR imaging biomarkers. As such, this work takes strides towards a pipeline with clinical utility, and the biomarkers identified further our understanding of OA progression and eventual TKR onset.

Multimodal Machine Learning-based Knee Osteoarthritis Progression Prediction from Plain Radiographs and Clinical Data

Knee osteoarthritis (OA) is the most common musculoskeletal disease without a cure, and current treatment options are limited to symptomatic relief. Prediction of OA progression is a very challenging and timely issue, and it could, if resolved, accelerate the disease modifying drug development and ultimately help to prevent millions of total joint replacement surgeries performed annually. Here, we present a multi-modal machine learning-based OA progression prediction model that utilises raw radiographic data, clinical examination results and previous medical history of the patient. We validated this approach on an independent test set of 3,918 knee images from 2,129 subjects. Our method yielded area under the ROC curve (AUC) of 0.79 (0.78–0.81) and Average Precision (AP) of 0.68 (0.66–0.70). In contrast, a reference approach, based on logistic regression, yielded AUC of 0.75 (0.74–0.77) and AP of 0.62 (0.60–0.64). The proposed method could significantly improve the subject selection process for OA drug-development trials and help the development of personalised therapeutic plans.

Thiazide diuretics and risk of knee replacement surgery among patients with knee osteoarthritis: a general population-based cohort study

OBJECTIVE

Thiazide diuretic use is associated with higher bone mineral density (BMD) and possibly lower serum magnesium levels than loop diuretic use, and both high BMD and low serum magnesium have been linked to high prevalent knee osteoarthritis. This study aimed to compare the risk of a clinically relevant endpoint, knee replacement (KR) surgery, among initiators of thiazide and loop diuretics.

DESIGN

Among patients aged ≥50 years with a diagnosis of knee osteoarthritis in The Health Improvement Network (THIN) in United Kingdom, we conducted a propensity score-matched cohort study to examine the relation of thiazide diuretic initiation vs loop diuretic initiation to the risk of KR over 5 years.

RESULTS

Among thiazide and loop diuretic initiators (n = 3,488 for each group; mean age: 73 years; female ratio: 59%), 359 (28.6/1,000 person-years) and 283 (24.1/1,000 person-years) KRs occurred during the follow-up period, respectively. The hazard ratio (HR) of KR for thiazide diuretic initiation vs loop diuretic initiation was 1.26 (95% confidence interval [CI]: 1.08-1.47). The adherence-adjusted HR of KR for continuous use of thiazide diuretics was 1.44 (95% CI: 1.21-1.72).

CONCLUSIONS

In this population-based cohort of patients with knee osteoarthritis, thiazide diuretic use was associated with a higher risk of KR than loop diuretic use. This association may potentially be due to thiazide diuretics' effect on BMD and serum magnesium.

Derivation and external validation of a risk prediction algorithm to estimate future risk of cardiovascular death among patients with type 2 diabetes and incident diabetic nephropathy: prospective cohort study

OBJECTIVE

To derive, and externally validate, a risk score for cardiovascular death among patients with type 2 diabetes and newly diagnosed diabetic nephropathy (DN).

RESEARCH DESIGN AND METHODS

Two independent prospective cohorts with type 2 diabetes were used to develop and externally validate the risk score. The derivation cohort comprised 2282 patients with an incident, clinical diagnosis of DN. The validation cohort includes 950 patients with incident, biopsy-proven diagnosis of DN. The outcome was cardiovascular death within 2 years of the diagnosis of DN. Logistic regression was applied to derive the risk score for cardiovascular death from the derivation cohort, which was externally validated in the validation cohort. The score was also estimated by applying the United Kingdom Prospective Diabetes Study (UKPDS) risk score in the external validation cohort.

RESULTS

The 2-year cardiovascular mortality was 12.05% and 11.79% in the derivation cohort and validation cohort, respectively. Traditional predictors including age, gender, body mass index, blood pressures, glucose, lipid profiles alongside novel laboratory test items covering five test panels (liver function, serum electrolytes, thyroid function, blood coagulation and blood count) were included in the final model.C-statistics was 0.736 (95% CI 0.731 to 0.740) and 0.747 (95% CI 0.737 to 0.756) in the derivation cohort and validation cohort, respectively. The calibration slope was 0.993 (95% CI 0.974 to 1.013) and 1.000 (95% CI 0.981 to 1.020) in the derivation cohort and validation cohort, respectively.The UKPDS risk score substantially underestimated cardiovascular mortality.

CONCLUSIONS

A new risk score based on routine clinical measurements that quantified individual risk of cardiovascular death was developed and externally validated. Compared with the UKPDS risk score, which underestimated the cardiovascular disease risk, the new score is a more specific tool for patients with type 2 diabetes and DN. The score could work as a tool to identify individuals at the highest risk of cardiovascular death among those with DN.