Explainable deep learning for disease activity prediction in chronic inflammatory joint diseases.

Analysing complex diseases such as chronic inflammatory joint diseases (CIJDs), where many factors influence the disease evolution over time, is a challenging task. CIJDs are rheumatic diseases that cause the immune system to attack healthy organs, mainly the joints. Different environmental, genetic and demographic factors affect disease development and progression. The Swiss Clinical Quality Management in Rheumatic Diseases (SCQM) Foundation maintains a national database of CIJDs documenting the disease management over time for 19'267 patients. We propose the Disease Activity Score Network (DAS-Net), an explainable multi-task learning model trained on patients' data with different arthritis subtypes, transforming longitudinal patient journeys into comparable representations and predicting multiple disease activity scores. First, we built a modular model composed of feed-forward neural networks, long short-term memory networks and attention layers to process the heterogeneous patient histories and predict future disease activity. Second, we investigated the utility of the model's computed patient representations (latent embeddings) to identify patients with similar disease progression. Third, we enhanced the explainability of our model by analysing the impact of different patient characteristics on disease progression and contrasted our model outcomes with medical expert knowledge. To this end, we explored multiple feature attribution methods including SHAP, attention attribution and feature weighting using case-based similarity. Our model outperforms temporal and non-temporal neural network, tree-based, and naive static baselines in predicting future disease activity scores. To identify similar patients, a k-nearest neighbours regression algorithm applied to the model's computed latent representations outperforms baseline strategies that use raw input features representation.

The integrin CD11b inhibits MSU-induced NLRP3 inflammasome activation in macrophages and protects mice against MSU-induced joint inflammation.

OBJECTIVE: In gout, monosodium urate crystals are taken up by macrophages, triggering the activation of the NLRP3 inflammasome and the maturation of IL-1ß. This study aimed to investigate the role of integrin CD11b in inflammasome activation in macrophages stimulated by MSU. METHODS: BMDM from WT and CD11b KO mice were stimulated in vitro with MSU crystals. Cellular supernatants were collected to assess the expression of the inflammatory cytokines by enzyme-linked immunosorbent assay and western blot methods. The role of integrin CD11b in MSU-induced gouty arthritis in vivo was investigated by intra-articular injection of MSU crystals. Real-time extracellular acidification rate and oxygen consumption rate of BMDMs were measured by Seahorse Extracellular Flux Analyzer. RESULTS: We demonstrate that CD11b-deficient mice developed exacerbated gouty arthritis with increased recruitment of leukocytes in the joint and higher IL-1ß levels in the sera. In macrophages, genetic deletion of CD11b induced a shift of macrophage metabolism from oxidative phosphorylation to glycolysis, thus decreasing the overall generation of intracellular ATP. Upon MSU stimulation, CD11b-deficient macrophages showed an exacerbated secretion of IL-1ß. Treating wild-type macrophages with a CD11b agonist, LA1, inhibited MSU-induced release of IL-1ß in vitro and attenuated the severity of experimental gouty arthritis. Importantly, LA1, was also effective in human cells as it inhibited MSU-induced release of IL-1ß by peripheral blood mononuclear cells from healthy donors. CONCLUSION: Our data identified the CD11b integrin as a principal cell membrane receptor that modulates NLRP3 inflammasome activation by MSU crystal in macrophages, which could be a potential therapeutic target to treat gouty arthritis in human patients.

Development of a Management App for Postviral Fibromyalgia-Like Symptoms: Patient Preference-Guided Approach.

BACKGROUND: Persistent fibromyalgia-like symptoms have been increasingly reported following viral infections, including SARS-CoV-2. About 30% of patients with post-COVID-19 syndrome fulfill the fibromyalgia criteria. This complex condition presents significant challenges in terms of self-management. Digital health interventions offer a viable means to assist patients in managing their health conditions. However, the challenge of ensuring their widespread adoption and adherence persists. This study responds to this need by developing a patient-centered digital health management app, incorporating patient preferences to enhance usability and effectiveness, ultimately aiming to improve patient outcomes and quality of life. OBJECTIVE: This research aims to develop a digital health self-management app specifically for patients experiencing postviral fibromyalgia-like symptoms. By prioritizing patient preferences and engagement through the app's design and functionality, the study intends to facilitate better self-management practices and improve adherence. METHODS: Using an exploratory study design, the research used patient preference surveys and usability testing as primary tools to inform the development process of the digital health solution. We gathered and analyzed patients' expectations regarding design features, content, and usability to steer the iterative app development. RESULTS: The study uncovered crucial insights from patient surveys and usability testing, which influenced the app's design and functionality. Key findings included a preference for a symptom list over an automated chatbot, a desire to report on a moderate range of symptoms and activities, and the importance of an intuitive onboarding process. While usability testing identified some challenges in the onboarding process, it also confirmed the importance of aligning the app with patient needs to enhance engagement and satisfaction. CONCLUSIONS: Incorporating patient feedback has been a significant factor in the development of the digital health app. Challenges encountered with user onboarding during usability testing have highlighted the importance of this process for user adoption. The study acknowledges the role of patient input in developing digital health technologies and suggests further research to improve onboarding procedures, aiming to enhance patient engagement and their ability to manage digital health resources effectively. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/32193.

Detection and Grading of Radiographic Hand Osteoarthritis Using an Automated Machine Learning Platform.

OBJECTIVE: Automated machine learning (autoML) platforms allow health care professionals to play an active role in the development of machine learning (ML) algorithms according to scientific or clinical needs. The aim of this study was to develop and evaluate such a model for automated detection and grading of distal hand osteoarthritis (OA). METHODS: A total of 13,690 hand radiographs from 2,863 patients within the Swiss Cohort of Quality Management (SCQM) and an external control data set of 346 non-SCQM patients were collected and scored for distal interphalangeal OA (DIP-OA) using the modified Kellgren/Lawrence (K/L) score. Giotto (Learn to Forecast [L2F]) was used as an autoML platform for training two convolutional neural networks for DIP joint extraction and subsequent classification according to the K/L scores. A total of 48,892 DIP joints were extracted and then used to train the classification model. Heatmaps were generated independently of the platform. User experience of a web application as a provisional user interface was investigated by rheumatologists and radiologists. RESULTS: The sensitivity and specificity of this model for detecting DIP-OA were 79% and 86%, respectively. The accuracy for grading the correct K/L score was 75%, with a κ score of 0.76. The accuracy per DIP-OA class differed, with 86% for no OA (defined as K/L scores 0 and 1), 71% for a K/L score of 2, 46% for a K/L score of 3, and 67% for a K/L score of 4. Similar values were obtained in an independent external test set. Qualitative and quantitative user experience testing of the web application revealed a moderate to high demand for automated DIP-OA scoring among rheumatologists. Conversely, radiologists expressed a low demand, except for the use of heatmaps. CONCLUSION: AutoML platforms are an opportunity to develop clinical end-to-end ML algorithms. Here, automated radiographic DIP-OA detection is both feasible and usable, whereas grading among individual K/L scores (eg, for clinical trials) remains challenging.

[Electronic health record : challenges and possibilities]. / Dossier patient informatisé : défis et possibilités.

Information systems have been driving technological advances in healthcare, especially over the past two decades. Digital health is transforming care by emphasizing efficiency, accessibility, and integrating technology at every stage of the patient journey. The Electronic Health Record (EHR), a software used in hospitals or clinics, emerges as a cornerstone in this transformation by consolidating medical data to facilitate care coordination. This integration promises to enhance the quality of care and the patient-healthcare professional relationship, while presenting practical challenges such as increased documentation. In this article, we explore various aspects of the EHR, addressing its challenges and potential adaptations in our practice.

Les systèmes d'information ont été des moteurs d'avancées technologiques en santé, surtout au cours des deux dernières décennies. La santé numérique révolutionne les soins en mettant l'accent sur l'efficacité et l'accessibilité, intégrant la technologie à chaque étape du parcours du patient. Le dossier patient informatisé (DPI), logiciel utilisé en milieu hospitalier ou dans un cabinet, émerge comme pivot de cette transformation en consolidant les données médicales pour faciliter la coordination des soins. Cette intégration promet d'améliorer la qualité des soins et la relation patient-professionnel de santé tout en posant des défis pratiques tels que la documentation accrue. Dans cet article, nous explorons divers aspects du DPI, abordant ses défis et les adaptations possibles dans notre pratique.

Advancing Rheumatology Care Through Machine Learning.

Rheumatologic diseases are marked by their complexity, involving immune-, metabolic- and mechanically mediated processes which can affect different organ systems. Despite a growing arsenal of targeted medications, many rheumatology patients fail to achieve full remission. Assessing disease activity remains challenging, as patients prioritize different symptoms and disease phenotypes vary. This is also reflected in clinical trials where the efficacy of drugs is not necessarily measured in an optimal way with the traditional outcome assessment. The recent COVID-19 pandemic has catalyzed a digital transformation in healthcare, embracing telemonitoring and patient-reported data via apps and wearables. As a further driver of digital medicine, electronic medical record (EMR) providers are actively engaged in developing algorithms for clinical decision support, heralding a shift towards patient-centered, decentralized care. Machine learning algorithms have emerged as valuable tools for handling the increasing volume of patient data, promising to enhance treatment quality and patient well-being. Convolutional neural networks (CNN) are particularly promising for radiological image analysis, aiding in the detection of specific lesions such as erosions, sacroiliitis, or osteoarthritis, with several FDA-approved applications. Clinical predictions, including numerical disease activity forecasts and medication choices, offer the potential to optimize treatment strategies. Numeric predictions can be integrated into clinical workflows, allowing for shared decision making with patients. Clustering patients based on disease characteristics provides a personalized care approach. Digital biomarkers, such as patient-reported outcomes and wearables data, offer insights into disease progression and therapy response more flexibly and outside patient consultations. In association with patient-reported outcomes, disease-specific digital biomarkers via image recognition or single-camera motion capture enables more efficient remote patient monitoring. Digital biomarkers may also play a major role in clinical trials in the future as continuous, disease-specific outcome measurement facilitating decentralized studies. Prediction models can help with patient selection in clinical trials, such as by predicting high disease activity. Efforts are underway to integrate these advancements into clinical workflows using digital pathways and remote patient monitoring platforms. In summary, machine learning, digital biomarkers, and advanced imaging technologies hold immense promise for enhancing clinical decision support and clinical trials in rheumatology. Effective integration will require a multidisciplinary approach and continued validation through prospective studies.

Inhibiting Lysyl Oxidases prevents pathologic cartilage calcification.

Lysyl oxidases (LOX(L)) are enzymes that catalyze the formation of cross-links in collagen and elastin fibers during physiologic calcification of bone. However, it remains unknown whether they may promote pathologic calcification of articular cartilage, an important hallmark of debilitating arthropathies. Here, we have studied the possible roles of LOX(L) in cartilage calcification, related and not related to their cross-linking activity. We first demonstrated that inhibition of LOX(L) by ß-aminoproprionitrile (BAPN) significantly reduced calcification in murine and human chondrocytes, and in joint of meniscectomized mice. These BAPN's effects on calcification were accounted for by different LOX(L) roles. Firstly, reduced LOX(L)-mediated extracellular matrix cross-links downregulated Anx5, Pit1 and Pit2 calcification genes. Secondly, BAPN reduced collagen fibrotic markers Col1 and Col3. Additionally, LOX(L) inhibition blocked chondrocytes hypertrophic differentiation (Runx2 and COL10), pro-inflammatory IL-6 release and reactive oxygen species (ROS) production, all triggers of chondrocyte calcification. Through unbiased transcriptomic analysis we confirmed a positive correlation between LOX(L) genes and genes for calcification, hypertrophy and extracellular matrix catabolism. This association was conserved throughout species (mouse, human) and tissues that can undergo pathologic calcification (kidney, arteries, skin). Overall, LOX(L) play a critical role in the process of chondrocyte calcification and may be therapeutic targets to treat cartilage calcification in arthropathies.

Elevated secretion of pro-collagen I-alpha and vascular endothelial growth factor as biomarkers of acetabular labrum degeneration and calcification in hip osteoarthritis: An explant study.

Background: Hip osteoarthritis (OA) involves structural degeneration of different joint compartments, including femoral head cartilage, periarticular ligaments and the acetabular labrum. However, the molecular mechanisms underlying labrum degeneration in hip OA remain poorly understood. Aim: To assess secretion of putative biomarkers for OA from explanted human labrum tissues under basal and inflammatory conditions and to determine whether these could differentiate between OA and calcification status compared to fracture controls. Methods: Intact labrum specimens were collected from patients undergoing joint arthroplasty for primary hip OA (n â€‹= â€‹15, mean age 70) or non-OA femoral neck fracture (n â€‹= â€‹5, mean age 64). Tissues were dissected in equal-sized samples and explanted for one week. To mimic activation of inflammatory signaling by endogenous damage-associated molecular patterns (DAMP) tissue were stimulated with a toll-like receptor 4 (TLR4) agonist (1 â€‹µg/mL LPS). The involvement of transforming growth factor-beta (TGF-beta) signaling was evaluated by treatment with a TGF-beta type 1 receptor inhibitor (10 â€‹µM SB-505124). Secretion of aggrecan (ACAN), pro-collagen-I alpha (Pro-Col-Iα), cartilage oligomeric matrix protein (COMP), interleukin-6 (IL-6) and vascular endothelial growth factor (VEGF) was assessed by enzyme-linked immunosorbent assay (ELISA). Labrum calcification was evaluated by 3D whole mount fluorescent microscopy of ethyl cinnamate-based optically cleared tissues stained with Alcian blue/Alizarin red. Results: Whole mount microscopy revealed non-OA fracture controls were non-calcified, whereas six OA labra (40%) were partially calcified or ossified. Basal secretion of Pro-Col-Iα and VEGF was increased four-fold in OA versus non-OA labra. Pro-Col-Iα levels were correlated with those of VEGF (r â€‹= â€‹0.65) and COMP (r â€‹= â€‹0.54). Stimulation of DAMP signaling through TLR4 affected secretion of IL-6, VEGF, COMP and Pro-Col-Iα, with distinct responses between non-OA and OA tissues. Inhibition of TGF-beta signaling specifically reduced elevated secretion of Pro-Col- Iα and VEGF in calcified OA labrum. Conclusions: Secretion of the putative OA biomarkers Pro-Col-Iα and VEGF is elevated in degenerated human acetabular labrum and may serve as indicators of OA and calcification status. Secretion of both factors was partially regulated by TGF-beta signaling in calcified OA labrum tissues.The Translational potential of this article:Our findings suggest that a biomarker panel consisting of Pro-Col-Iα/VEGF/COMP may be valuable for assessing subradiographic labrum degeneration and calcification in hip OA. Targeting TGF-beta signaling may offer a means to reduce vascular invasion and fibrosis in acetabular labrum tissue.

A new ex vivo human model of osteoarthritis cartilage calcification.

OBJECTIVE: Cartilage pathologic calcification is a hallmark of osteoarthritis (OA). Here, we aimed to describe a new ex vivo human model to study the progression of cartilage calcification. METHOD: Cartilage explants (n = 11), as well as primary chondrocytes (n = 3), were obtained from OA patients undergoing knee replacement. Explants and chondrocytes were cultured in control (NT) or calcification (CM) medium (supplemented with ascorbic acid and ß-glycerophosphate). Calcification was evaluated by micro-CT scan at day 0 and 21 in explants, and by Alizarin red staining in chondrocyte monolayers. Raman spectrometry allowed characterization of the crystal type. Interleukin-6 (IL-6) secretion in explant and cell supernatants was measured by ELISA. Finally, matrix degradation was evaluated by Safranin-O staining of explant sections and by glycosaminoglycans (GAG) release in supernatants. RESULTS: Micro-CT scan showed calcifications in all explants at baseline (day 0), which in the CM group increased significantly in number and size after 21 days compared with the NT group. Raman spectrometry revealed that crystals were exclusively basic calcium phosphate crystals (carbonated hydroxyapatite) both in NT and CM. IL-6 secretion was significantly increased in calcifying conditions. Finally, CM significantly increased cartilage catabolism as assessed by decreased Safranin-O staining of tissue explants and increased GAG release in supernatants. CM effects (enhanced calcification, IL-6 secretion and proteoglycans turn-over) were recapitulated in vitro in OA chondrocytes. CONCLUSIONS: We have described a new ex vivo human model of cartilage calcification that can summurize the triad of events seen during osteoarthritis progression, i.e. calcification, inflammation, and cartilage degradation. This model will allow the identification of new anti-calcification compounds.

Early axial spondyloarthritis according to the ASAS consensus definition: characterisation of patients and effectiveness of a first TNF inhibitor in a large observational registry.

OBJECTIVE: To characterise the population fulfilling the Assessment of SpondyloArthritis international Society (ASAS) consensus definition of early axial spondyloarthritis (axSpA) and to determine the effectiveness of a first tumour necrosis factor inhibitor (TNFi) in early versus established axSpA in a large observational registry. METHODS: A total of 3064 patients with axSpA in the Swiss Clinical Quality Management registry with data on duration of axial symptoms were included (≤2 years=early axSpA, N=658; >2 years=established axSpA, N=2406). Drug retention was analysed in patients starting a first TNFi in early axSpA (N=250) versus established axSpA (N=874) with multiple-adjusted Cox proportional hazards models. Adjusted logistic regression analyses were used to determine the achievement of the ASAS criteria for 40% improvement (ASAS40) at 1 year. RESULTS: Sex distribution, disease activity, impairments of function and health-related quality of life were comparable between patients with early and established axSpA. Patients with established disease were older, had more prevalent axial radiographical damage and had a higher impairment of mobility. A comparable TNFi retention was found in early versus established disease after adjustment for age, sex, human leucocyte antigen-B27 status, education, body mass index, smoking, elevated C reactive protein and sacroiliac inflammation on MRI (HR 1.05, 95% CI 0.78 to 1.42). The adjusted ASAS40 response was similar in the two groups (OR 1.09, 95% CI 0.67 to 1.78). Results were confirmed in the population fulfilling the ASAS classification criteria. CONCLUSION: Considering the recent ASAS definition of early axSpA, TNFi effectiveness seems comparable in early versus established disease.

Digital transformation of an academic hospital department: A case study on strategic planning using the balanced scorecard.

Digital transformation has a significant impact on efficiency and quality in hospitals. New solutions can support the management of data overload and the shortage of qualified staff. However, the timely and effective integration of these new digital tools in the healthcare setting poses challenges and requires guidance. The balanced scorecard (BSC) is a managerial method used to translate new strategies into action and measure their impact in an institution, going beyond financial values. This framework enables quicker operational adjustments and enhances awareness of real-time performance from multiple perspectives, including customers, internal procedures, and the learning organization. The aim of this study was to adapt the BSC to the evolving digital healthcare environment, encompassing factors like the recent pandemic, new technologies such as artificial intelligence, legislation, and user preferences. A strategic mapping with identification of corresponding key performance indicators was performed. To achieve this, we employed a qualitative research approach involving retreats, interdisciplinary working groups, and semi-structured interviews with different stakeholders (administrative, clinical, computer scientists) in a rheumatology department. These inputs served as the basis for customizing the BSC according to upcoming or already implemented solutions and to define actionable, cross-level performance indicators for all perspectives. Our defined values include quality of care, patient empowerment, employee satisfaction, sustainability and innovation. We also identified substantial changes in our internal processes, with the electronic medical record (EMR) emerging as a central element for vertical and horizontal digitalization. This includes integrating patient-reported outcomes, disease-specific digital biomarker, prediction algorithms to increase the quality of care as well as advanced language models in order save resources. Gaps in communication and collaboration between medical departments have been identified as a main target for new digital solutions, especially in patients with more than one disorder. From a learning institution's perspective, digital literacy among patients and healthcare professionals emerges as a crucial lever for successful implementation of internal processes. In conclusion, the BSC is a helpful tool for guiding digitalization in hospitals as a horizontally and vertically connected process that affects all stakeholders. Future studies should include empirical analyses and explore correlations between variables and above all input and user experience from patients.

Longitudinal associations between body mass index and changes in disease activity and radiographic progression in rheumatoid arthritis patients treated with infliximab.

OBJECTIVES: Treatment response is worse in obese patients with rheumatoid arthritis (RA), including patients on weight-adjusted therapies like infliximab. We aimed to assess the association between body mass index (BMI) and changes in RA disease activity and radiographic progression over time. METHODS: We included infliximab users with an RA diagnosis in the Swiss Clinical Quality Management in Rheumatic Diseases registry (1997-2020). Two cohorts were defined: (1) starting from their first BMI measurement or disease activity score (DAS28-esr), and (2) from their first BMI measurement or radiographic assessment (Rau score). We evaluated the coefficient and 95% CI of BMI with changes in mean DAS28-esr (cohort 1) and mean Rau scores (a structural joint damage score, cohort 2) using generalised estimation equations, overall and stratified by BMI categories. RESULTS: Cohort 1 comprised 412 patients (74% women, mean age 53 years, mean BMI 25). We observed no change in mean DAS28-esr with increasing BMI overall (adjusted coefficient: 0.00, 95% CI -0.02 to 0.02), or in BMI categories. Cohort 2 comprised 187 patients highly alike to those in cohort 1. We observed a significant decrease of 1.05 in mean Rau scores for every increase in BMI unit (adjusted coefficient: -1.05, 95% CI -1.92 to -0.19). Results remained statistically non-significant across BMI categories. CONCLUSIONS: Our longitudinal investigation suggests that BMI increase may not lead to changes in DAS28-esr in patients receiving infliximab, despite the weight-adapted dose. Yet, there may be a decrease in erosions with increasing weight non-limited to obese patients.

Real-life drug retention rate and safety of rituximab when treating rheumatic diseases: a single-centre Swiss retrospective cohort study.

BACKGROUND: In Switzerland, rituximab (RTX) is licenced for the treatment of rheumatoid arthritis (RA) and ANCA-associated vasculitis (AAV) but is frequently used off-label to treat other auto-immune diseases (AID), especially connective tissue diseases (CTD). We aimed to characterise the use of RTX in AID in a real-life Swiss setting and compare RTX retention rates and safety outcomes between patients treated for RA, CTD and AAV. METHODS: A retrospective cohort study of patients who started RTX in the Rheumatology Department for RA or AID. The RTX retention rate was analysed using Kaplan-Meier survival curves. Occurrences of serious adverse events (SAE), low IgG levels and anti-drug antibodies (ADA) were reported. RESULTS: Two hundred three patients were treated with RTX: 51.7% had RA, 29.6% CTD, 9.9% vasculitis and 8.9% other AIDs. The total observation time was 665 patient-years. RTX retention probability at 2 years (95%CI) was similar for RA and CTD 0.65 (0.55 to 0.73), 0.60 (0.47 to 0.72) and lower for vasculitis 0.25 (0.09 to 0.45). Survival curves for RTX retention matched closely (p = 0.97) between RA and CTD patients but were lower for patients with vasculitis due to a higher percentage of induced remission. Patients with vasculitis (95%) and CTD (75%) had a higher rate of concomitant glucocorticoid use than RA (60%). Moderate to severe hypogammaglobulinaemia was observed more frequently in patients with vasculitis (35%) than with RA (13%) or CTD (9%) and was associated with an increased risk of presenting a first infectious SAE (HR 2.01, 95% CI 1.04 to 3.91). The incidence rate of SAE was 23.3 SAE/100 patient-years (36% were infectious). When searched, ADAs were observed in 18% of the patients and were detected in 63% of infusions-related SAE. 10 patients died during RTX treatment and up to 12 months after the last RTX infusion, 50% from infection. CONCLUSION: RTX retention rates are similar for patients with RA and CTD but lower for those with vasculitis due to more frequent remission. Patients treated with RTX for vasculitis present more SAE and infectious SAE than patients with RA and CTD, potentially due to a higher use of concomitant glucocorticoids and the occurrence of hypogammaglobulinaemia.

Patient groups in Rheumatoid arthritis identified by deep learning respond differently to biologic or targeted synthetic DMARDs.

Cycling of biologic or targeted synthetic disease modifying antirheumatic drugs (b/tsDMARDs) in rheumatoid arthritis (RA) patients due to non-response is a problem preventing and delaying disease control. We aimed to assess and validate treatment response of b/tsDMARDs among clusters of RA patients identified by deep learning. We clustered RA patients clusters at first-time b/tsDMARD (cohort entry) in the Swiss Clinical Quality Management in Rheumatic Diseases registry (SCQM) [1999-2018]. We performed comparative effectiveness analyses of b/tsDMARDs (ref. adalimumab) using Cox proportional hazard regression. Within 15 months, we assessed b/tsDMARD stop due to non-response, and separately a ≥20% reduction in DAS28-esr as a response proxy. We validated results through stratified analyses according to most distinctive patient characteristics of clusters. Clusters comprised between 362 and 1481 patients (3516 unique patients). Stratified (validation) analyses confirmed comparative effectiveness results among clusters: Patients with ≥2 conventional synthetic DMARDs and prednisone at b/tsDMARD initiation, male patients, as well as patients with a lower disease burden responded better to tocilizumab than to adalimumab (hazard ratio [HR] 5.46, 95% confidence interval [CI] [1.76-16.94], and HR 8.44 [3.43-20.74], and HR 3.64 [2.04-6.49], respectively). Furthermore, seronegative women without use of prednisone at b/tsDMARD initiation as well as seropositive women with a higher disease burden and longer disease duration had a higher risk of non-response with golimumab (HR 2.36 [1.03-5.40] and HR 5.27 [2.10-13.21], respectively) than with adalimumab. Our results suggest that RA patient clusters identified by deep learning may have different responses to first-line b/tsDMARD. Thus, it may suggest optimal first-line b/tsDMARD for certain RA patients, which is a step forward towards personalizing treatment. However, further research in other cohorts is needed to verify our results.

The associations between bariatric surgery and hip or knee arthroplasty, and hip or knee osteoarthritis: Propensity score-matched cohort studies.

Objective: To investigate the associations between bariatric surgery and hip or knee arthroplasty, and secondary care hip or knee osteoarthritis (OA). Methods: We performed cohort studies using data from Swedish nationwide healthcare registries. Patients aged 18-79 years who underwent bariatric surgery between 2006 and 2019 were matched on their propensity score (PS) to up to 2 obese patients ("unexposed episodes") in risk-set sampling. After a 1-year run-in period, episodes were followed in an "as-treated" approach. Using Cox proportional hazard regression, we calculated hazard ratios (HR) with 95% confidence intervals (CIs) of hip or knee arthroplasty overall and in subgroups of age, sex, joint location, arthroplasty type, bariatric surgery type, and by duration of follow-up if proportional hazard assumptions were violated. In a secondary cohort, we assessed the outcome incident secondary care hip or knee osteoarthritis (OA). Results: Among 39'392 bariatric surgery episodes when compared to 61'085 â€‹PS-matched unexposed episodes (47'594 unique patients), the risk of hip or knee arthroplasty was strongest increased within the first three years of follow-up (HR 1.79, 95% CI 1.56-2.07), decreased thereafter, but remained elevated throughout follow-up. In a secondary cohort of 37'929 exposed when compared to 58'600 â€‹PS-matched unexposed episodes, the risk of hip or knee osteoarthritis was decreased (HR 0.84, 95% CI 0.79-0.90). Conclusion: Bariatric surgery is associated with increased risks of hip or knee arthroplasty, but also with decreased risks of secondary care OA. This contradiction supports the hypothesis that bariatric surgery may act as an enabler for hip or knee arthroplasty.

Polg mtDNA mutator mice reveal limited involvement of vertebral bone loss in premature aging-related thoracolumbar hyperkyphosis.

Background: Age-related hyperkyphosis is multifactorial and involves alterations of vertebral bone, intervertebral discs (IVD) and paraspinal muscles. The relative contribution of these tissues remains unclear. Here, we compared differences in vertebral bone microarchitecture and IVD thickness between prematurely aging mice with spinal hyperkyphosis and wild type littermates. Methods: Thoracolumbar vertebral columns were dissected from homozygous Polg D257A and age-matched wild type littermates. Micro-computed tomography was performed to quantify cortical and trabecular bone parameters at anterior and posterior portions of T8-L4 vertebrae. In addition, vertebral shape, transaxial facet joint orientation and IVD thickness were quantified. Differences in anterior/posterior ratios between genotypes were compared by Student's t-test and association between vertebral bone and IVD parameters was investigated using Pearson correlation analysis. Results: Hyperkyphotic homozygous mice displayed generalized osteopenia that was more pronounced at the posterior compared with anterior portion of thoracolumbar vertebrae. An increase in the anterior/posterior ratio of trabecular bone parameters was revealed at the thoracolumbar junction (T13-L1). Polg D257A displayed diffuse loss of cortical bone thickness, yet anterior/posterior ratios were unchanged. Despite generalized and regional bone loss, vertebral shape was unaffected. PolG D257A mice showed a 10-20 % reduction of IVD thickness at both thoracic and lumbar levels, with only minimal histopathological changes. IVD thickness was negatively correlated with anterior/posterior ratios of trabecular bone parameters, as well as with more coronally oriented facet joints, but negatively correlated with the anterior/posterior ratio of cortical bone thickness. Conclusions: Aging-induced regional changes of vertebral trabecular and cortical bone did not lead to altered vertebral shape in Polg D257A mice but may indirectly cause hyperkyphosis through reduction of IVD thickness. These findings suggest a limited role for aging-induced bone loss in spinal hyperkyphosis and warrants further research on the involvement of paraspinal muscle degeneration.