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

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

Glucagon resistance in individuals with obesity and hepatic steatosis can be measured using the GLUSENTIC test and index.

Increased plasma levels of glucagon (hyperglucagonaemia) promote diabetes development but is also observed in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). This may reflect hepatic glucagon resistance towards amino acid catabolism. A clinical test for measuring glucagon resistance has not been validated. We evaluated our glucagon sensitivity (GLUSENTIC) test, consisting of two study days: a glucagon injection and measurements of plasma amino acids, and an infusion of mixed amino acids and subsequent calculation of the GLUSENTIC index (primary outcome measure) from measurements of glucagon and amino acids. To distinguish glucagon-dependent from insulin-dependent actions on amino acid metabolism, we also studied patients with type 1 diabetes (T1D). The delta-decline in total amino acids was 49% lower in MASLD following exogenous glucagon (p=0.01), and the calculated GLUSENTIC index was 34% lower in MASLD (p<0.0001), but not T1D (p>0.99). In contrast, glucagon-induced glucose increments were similar in controls and MASLD (p=0.41). The GLUSENTIC test and index may be used to measure glucagon resistance in individuals with obesity and MASLD.

CT in musculoskeletal imaging: still helpful and for what?

Computed tomography (CT) is a common modality employed for musculoskeletal imaging. Conventional CT techniques are useful for the assessment of trauma in detection, characterization and surgical planning of complex fractures. CT arthrography can depict internal derangement lesions and impact medical decision making of orthopedic providers. In oncology, CT can have a role in the characterization of bone tumors and may elucidate soft tissue mineralization patterns. Several advances in CT technology have led to a variety of acquisition techniques with distinct clinical applications. These include four-dimensional CT, which allows examination of joints during motion; cone-beam CT, which allows examination during physiological weight-bearing conditions; dual-energy CT, which allows material decomposition useful in musculoskeletal deposition disorders (e.g., gout) and bone marrow edema detection; and photon-counting CT, which provides increased spatial resolution, decreased radiation, and material decomposition compared to standard multi-detector CT systems due to its ability to directly translate X-ray photon energies into electrical signals. Advanced acquisition techniques provide higher spatial resolution scans capable of enhanced bony microarchitecture and bone mineral density assessment. Together, these CT acquisition techniques will continue to play a substantial role in the practices of orthopedics, rheumatology, metabolic bone, oncology, and interventional radiology.

Artificial intelligence for MRI stroke detection: a systematic review and meta-analysis.

OBJECTIVES: This systematic review and meta-analysis aimed to assess the stroke detection performance of artificial intelligence (AI) in magnetic resonance imaging (MRI), and additionally to identify reporting insufficiencies. METHODS: PRISMA guidelines were followed. MEDLINE, Embase, Cochrane Central, and IEEE Xplore were searched for studies utilising MRI and AI for stroke detection. The protocol was prospectively registered with PROSPERO (CRD42021289748). Sensitivity, specificity, accuracy, and area under the receiver operating characteristic (ROC) curve were the primary outcomes. Only studies using MRI in adults were included. The intervention was AI for stroke detection with ischaemic and haemorrhagic stroke in separate categories. Any manual labelling was used as a comparator. A modified QUADAS-2 tool was used for bias assessment. The minimum information about clinical artificial intelligence modelling (MI-CLAIM) checklist was used to assess reporting insufficiencies. Meta-analyses were performed for sensitivity, specificity, and hierarchical summary ROC (HSROC) on low risk of bias studies. RESULTS: Thirty-three studies were eligible for inclusion. Fifteen studies had a low risk of bias. Low-risk studies were better for reporting MI-CLAIM items. Only one study examined a CE-approved AI algorithm. Forest plots revealed detection sensitivity and specificity of 93% and 93% with identical performance in the HSROC analysis and positive and negative likelihood ratios of 12.6 and 0.079. CONCLUSION: Current AI technology can detect ischaemic stroke in MRI. There is a need for further validation of haemorrhagic detection. The clinical usability of AI stroke detection in MRI is yet to be investigated. CRITICAL RELEVANCE STATEMENT: This first meta-analysis concludes that AI, utilising diffusion-weighted MRI sequences, can accurately aid the detection of ischaemic brain lesions and its clinical utility is ready to be uncovered in clinical trials. KEY POINTS: There is a growing interest in AI solutions for detection aid. The performance is unknown for MRI stroke assessment. AI detection sensitivity and specificity were 93% and 93% for ischaemic lesions. There is limited evidence for the detection of patients with haemorrhagic lesions. AI can accurately detect patients with ischaemic stroke in MRI.

The Impact of Weight-bearing Exercise, Non-Weight-bearing Exercise, and Cardiovascular Stress on Biochemical Markers of Cartilage Turnover in Patients With Mild to Moderate Knee Osteoarthritis: A Sequential, Cross-Over, Clinical Study.

OBJECTIVE: To investigate how running, cycling, and sedentary cardiovascular stress impact biomarkers of cartilage turnover acutely in subjects with knee osteoarthritis (OA). DESIGN: This was a sequential, cross-over, clinical study. Forty subjects with primary knee OA underwent moderate-to-high-intensity cycling, running, and adrenaline infusion on separate days. Blood was sampled before, during, and at 6-time points after intervention. On a control day, similar samples were taken. Biomarkers of type II collagen degradation (C2M, T2CM, Coll2-1, Coll2-1NO2), formation (PRO-C2), and aggrecan degradation (ARGS) were measured. RESULTS: Mean age was 60.4 years, 40% were male, 45% had cumulated Kellgren-Lawrence (KL)-grade (Right + Left knee) of 2 to 3 and 55% had 4 to 6. Analyzing overall changes, area under the curve was significantly lower compared with resting values for ARGS and C2M after cycling and for ARGS after running. Considering individual time points, peak changes in biomarker levels showed reduction in C2M shortly following cycling (T20min = -12.3%, 95% confidence interval [CI]: -19.3% to -5.2%). PRO-C2 increased during cycling (T10min = 14.0%, 95% CI = 4.1% to 23.8%) and running (T20min = 16.5%, 95% CI = 4.3% to 28.6%). T2CM decreased after cycling (T50min = -19.9%, 95% CI = -29.2% to -10.6%), running (T50min = -22.8%, 95% CI = -32.1% to -13.5%), and infusion of adrenaline (peak, T50min = -9.8%, 95% CI = -20.0% to 0.4%). A latent increase was seen in Coll2-1 240 minutes after running (T260min = 21.7%, 95% CI = -1.6% to 45.1%). CONCLUSION: Exercise had an impact on cartilage markers, but it did not suggest any detrimental effect on cartilage. Changes following adrenaline infusion suggest a sympathomimetic influence on the serological composition of biomarkers.

Improving traumatic fracture detection on radiographs with artificial intelligence support: a multi-reader study.

Objectives: The aim of this study was to evaluate the diagnostic performance of nonspecialist readers with and without the use of an artificial intelligence (AI) support tool to detect traumatic fractures on radiographs of the appendicular skeleton. Methods: The design was a retrospective, fully crossed multi-reader, multi-case study on a balanced dataset of patients (≥2 years of age) with an AI tool as a diagnostic intervention. Fifteen readers assessed 340 radiographic exams, with and without the AI tool in 2 different sessions and the time spent was automatically recorded. Reference standard was established by 3 consultant radiologists. Sensitivity, specificity, and false positives per patient were calculated. Results: Patient-wise sensitivity increased from 72% to 80% (P < .05) and patient-wise specificity increased from 81% to 85% (P < .05) in exams aided by the AI tool compared to the unaided exams. The increase in sensitivity resulted in a relative reduction of missed fractures of 29%. The average rate of false positives per patient decreased from 0.16 to 0.14, corresponding to a relative reduction of 21%. There was no significant difference in average reading time spent per exam. The largest gain in fracture detection performance, with AI support, across all readers, was on nonobvious fractures with a significant increase in sensitivity of 11 percentage points (pp) (60%-71%). Conclusions: The diagnostic performance for detection of traumatic fractures on radiographs of the appendicular skeleton improved among nonspecialist readers tested AI fracture detection support tool showed an overall reader improvement in sensitivity and specificity when supported by an AI tool. Improvement was seen in both sensitivity and specificity without negatively affecting the interpretation time. Advances in knowledge: The division and analysis of obvious and nonobvious fractures are novel in AI reader comparison studies like this.

Glucagon augments the secretion of FGF21 and GDF15 in MASLD by indirect mechanisms.

INTRODUCTION: Glucagon receptor agonism is currently explored for the treatment of obesity and metabolic dysfunction-associated steatotic liver disease (MASLD). The metabolic effects of glucagon receptor agonism may in part be mediated by increases in circulating levels of Fibroblast Growth Factor 21 (FGF21) and Growth Differentiation Factor 15 (GDF15). The effect of glucagon agonism on FGF21 and GDF15 levels remains uncertain, especially in the context of elevated insulin levels commonly observed in metabolic diseases. METHODS: We investigated the effect of a single bolus of glucagon and a continuous infusion of glucagon on plasma concentrations of FGF21 and GDF15 in conditions of endogenous low or high insulin levels. The studies included individuals with overweight with and without MASLD, healthy controls (CON) and individuals with type 1 diabetes (T1D). The direct effect of glucagon on FGF21 and GDF15 was evaluated using our in-house developed isolated perfused mouse liver model. RESULTS: FGF21 and GDF15 correlated with plasma levels of insulin, but not glucagon, and their secretion was highly increased in MASLD compared with CON and T1D. Furthermore, FGF21 levels in individuals with overweight with or without MASLD did not increase after glucagon stimulation when insulin levels were kept constant. FGF21 and GDF15 levels were unaffected by direct stimulation with glucagon in the isolated perfused mouse liver. CONCLUSION: The glucagon-induced secretion of FGF21 and GDF15 is augmented in MASLD and may depend on insulin. Thus, glucagon receptor agonism may augment its metabolic benefits in patients with MASLD through enhanced secretion of FGF21 and GDF15.

Long-term efficacy of a 2-year MRI treat-to-target strategy on disease activity and radiographic progression in patients with rheumatoid arthritis in clinical remission: 5-year follow-up of the IMAGINE-RA randomised trial.

OBJECTIVE: To investigate whether a 2-year MRI treat-to-target strategy targeting the absence of osteitis combined with clinical remission, compared with a conventional treat-to-target strategy targeting clinical remission only (IMAGINE-rheumatoid arthritis (RA) trial) improves clinical and radiographic outcomes over 5 years in patients with RA in clinical remission. METHODS: IMAGINE-more was an observational extension study of the original 2-year IMAGINE-RA randomised trial (NCT01656278). Clinical examinations and radiographs (hands and feet) were obtained yearly. Prespecified coprimary outcomes at year 5 were Disease Activity Score in 28 joints C reactive protein (DAS28-CRP) remission rate (DAS28-CRP<2.6) and no radiographic progression (van der Heijde-modified Sharp score (vdHSS) ≤0) from baseline. Secondary outcomes included 5-year changes in radiographic, MRI and clinical measures of disease activity and physical function. RESULTS: In total 131 patients, 86 women (67%), mean age 61.2, disease duration 9.5 years, median baseline DAS28-CRP 1.9 (IQR 1.6-2.2) and vdHSS 16.0 (IQR 7.0-36.0) were included in the study; 59 (59%) patients from the original MRI treat-to-target group and 72 (72%) from the conventional group. At year 5, 47 patients (80%) in the MRI treat-to-target group vs 54 patients (75%) in the conventional treat-to-target group were in DAS28-CRP remission (OR 2.00 (95% CI 0.76 to 5.28); p=0.16) while 14 patients (24%) vs 19 patients (26%) had no radiographic progression (OR 0.70, (95% CI 0.28 to 1.71); p=0.43). CONCLUSION: A 2-year combined MRI and clinical treat-to-target strategy, compared with a conventional clinical treat-to-target strategy alone, had no effect on the long-term probability of achieving DAS28-CRP remission and of avoiding radiographic progression.

Effect of repeated prolonged exercise on liver fat content and visceral adipose tissue in well-trained older men.

INTRODUCTION: Liver fat (LF) and visceral adipose tissue (VAT) content decreases with training, however, this has mainly been investigated in sedentary obese or healthy participants. The aim of this study was to investigate the effects of repeated prolonged exercise on LF and VAT content in well-trained older men and to compare baseline LF and VAT content to recreationally active older men. METHOD: A group of five well-trained older men were tested before and after cycling a total distance of 2558 km in 16 consecutive days. VAT content and body composition was measured using DXA before a bicycle ergometer test was performed to determine maximal fat oxidation (MFO), maximal oxygen consumption ( VO 2 max $$ {\mathrm{VO}}_{2_{\mathrm{max}}} $$ ), and the relative intensity at which MFO occurred (Fatmax). LF content was measured on a separate day using MRI. For comparison of baseline values, a control group of eight healthy age- and BMI-matched recreationally active men were recruited. RESULTS: The well-trained older men had lower VAT (p = 0.02), and a tendency toward lower LF content (p = 0.06) compared with the control group. The intervention resulted in decreased LF content (p = 0.02), but VAT, fat mass, and lean mass remained unchanged. VO 2 max $$ {\mathrm{VO}}_{2_{\mathrm{max}}} $$ , MFO, and Fatmax were not affected by the intervention. CONCLUSION: The study found that repeated prolonged exercise reduced LF content, but VAT and VO 2 max $$ {\mathrm{VO}}_{2_{\mathrm{max}}} $$ remained unchanged. Aerobic capacity was aligned with lower LF and VAT in older active men.

Tendon Elongation and Function After Delayed or Standard Loading of Surgically Repaired Achilles Tendon Ruptures: A Randomized Controlled Trial.

BACKGROUND: Achilles tendon ruptures often result in long-term functional deficits despite accelerated (standard) rehabilitation. PURPOSE/HYPOTHESIS: The purpose of this study was to investigate if delayed loading would influence functional, clinical, and structural outcomes of the muscles and tendon 1 year after a surgical repair. It was hypothesized that delaying the loading would reduce the heel-rise height deficit 1 year after Achilles tendon rupture. STUDY DESIGN: Randomized controlled trial; Level of evidence, 1. METHODS: In total, 48 patients with a surgically repaired Achilles tendon rupture were randomized to 2 groups: the standard group received the currently accepted rehabilitation, and the delayed group received the same rehabilitation except that initial loading was delayed by 6 weeks. The primary outcome was the heel-rise height difference between the injured and uninjured sides at 1 year. The secondary outcomes were (1) tendon length measured with magnetic resonance imaging, (2) muscle fascicle length and pennation angle of the gastrocnemius medialis muscle, (3) Doppler activity measured with ultrasonography, (4) Achilles tendon Total Rupture Score (ATRS), and (5) isometric muscle strength. RESULTS: The mean heel-rise height deficits for the standard and delayed groups were -2.2 cm and -2.1 cm, respectively (P = .719). The soleus part of the tendon was already elongated 1 week after surgery in both groups without a between-group difference (side-to-side difference: standard, 16.3 mm; delayed, 17.5 mm; P = .997) and did not change over 52 weeks. The gastrocnemius tendon length was unchanged at 1 week but elongated over time without a between-group difference (side-to-side difference at 52 weeks: standard, 10.5 mm; delayed, 13.0 mm; P = .899). The delayed group had less Doppler activity at 12 weeks (P = .006) and a better ATRS (standard, 60 points; delayed, 72 points; P = .032) at 52 weeks. CONCLUSION: Delayed loading was not superior to standard loading in reducing the heel-rise height difference at 1 year. The data indirectly suggested reduced inflammation in the initial months and a better patient-reported outcome at 1 year in the delayed group. The soleus part of the tendon was already markedly elongated (35%) 1 week after surgery, while the length of the gastrocnemius tendon was unchanged at 1 week but was 6% elongated at 1 year. Together, these data indirectly suggest that the delayed group fared better, although this finding needs to be confirmed in future investigations. REGISTRATION: NCT04263493 (ClinicalTrials.gov identifier).

Weight Loss for Patients With Gout and Concomitant Obesity: A Proof-of-Concept Randomized Trial.

OBJECTIVE: Despite scarce evidence, guidelines recommend weight loss as a management strategy for patients with gout. We investigated the effect of an intensive dietary intervention on body weight and clinical measures of gout severity in individuals with obesity and gout. METHODS: We conducted a 16-week randomized nonmasked parallel-group trial in Denmark, randomly assigning (one-to-one) individuals with obesity and gout to a low-energy diet or a control diet. The primary outcome was change in body weight. Key secondary outcomes were changes in serum urate (SU) level and visual analog scale-assessed pain and fatigue. RESULTS: Between December 1, 2018, and June 1, 2019, 61 participants were included in the intention-to-treat population and randomly assigned to the intensive diet group (n = 29) or control diet group (n = 32). Participants had a mean age of 60.3 (SD 9.9) years and mean body mass index of 35.6 (SD 5.0), and 59 (97%) were men. After 16 weeks, there was a significant difference in change in body weight between the diet and control groups (-15.4 vs -7.7 kg; difference -7.7 kg [95% confidence interval -10.7 to -4.7], P < 0.001). Despite results being potentially in favor of a low-energy diet, we could not confirm differences in SU level changes and fatigue between groups. No differences in pain and gout flares were observed between groups. No serious adverse events or deaths occurred during the trial. CONCLUSION: An intensive dietary intervention was safe and effectively lowered body weight in people with obesity and gout, but the weight loss did not directly translate into effects on SU level, fatigue, and pain.

Bisphosphonate Use and Risk of Atypical Femoral Fractures: A Danish Case Cohort Study with Blinded Radiographic Review.

CONTEXT: Prolonged bisphosphonate (BP) treatment for osteoporosis prevents hip and other fractures but causes atypical femoral fractures (AFF). OBJECTIVE: To establish the relationship between patterns of BP use and the risk of AFF and hip fractures. Other potential risk factors for AFF were also examined. DESIGN: Population-based case-cohort study. SETTING: The Danish National Healthcare system maintains longitudinal records of medication use, healthcare utilization, and x-ray images. PARTICIPANTS: Among all 1.9 million Danish adults ≥50, those with subtrochanteric or femoral shaft fractures between 2010-2015 (n = 4,973) were identified and compared to a random sample (n = 37,021). PREDICTORS: Bisphosphonate use was collected from 1995-2015. MAIN OUTCOME MEASURES: Fracture radiographs (n = 4,769) were reviewed by blinded study radiologists to identify AFFs (n = 181) using established criteria. Traditional hip fractures in the random sample (n = 691) were identified by ICD-10. RESULTS: Compared to <1 year of BP use, 5-7 years of use was associated with a 7-fold increase in AFF [adjusted HR = 7.29 (CI: 3.07,17.30)]; the risk of AFF fell quickly after discontinuation. The 5-year number-needed-to-harm for one AFF was 1,424, while the 5-year number-needed-to-treat to prevent one hip fracture was 56. Glucocorticoid and proton pump inhibitor use were independently associated with increased AFF risk. Thirty-one percent of those with AFF had no BP exposure. CONCLUSIONS: The risk of AFF increases with duration of BP use but the beneficial effects of BP therapy in adults ≥50 dramatically exceed this increased risk. Nearly one-third of those with AFF have no BP exposure.

Computed tomography or chest X-ray to assess pulmonary congestion in dyspnoeic patients with acute heart failure.

AIMS: While computed tomography (CT) is widely acknowledged as superior to chest radiographs for acute diagnostics, its efficacy in diagnosing acute heart failure (AHF) remains unexplored. This prospective study included consecutive patients with dyspnoea undergoing simultaneous low-dose chest CT (LDCT) and chest radiographs. Here, we aimed to determine if LDCT is superior to chest radiographs to confirm pulmonary congestion in dyspnoeic patients with suspected AHF. METHODS AND RESULTS: An observational, prospective study, including dyspnoeic patients from the emergency department. All patients underwent concurrent clinical examination, laboratory tests, echocardiogram, chest radiographs, and LDCT. The primary efficacy measure to compare the two radiological methods was conditional odds ratio (cOR). The primary outcome was adjudicated AHF, ascertained by comprehensive expert consensus. The secondary outcome, echo-bnp AHF, was an objective AHF diagnosis based on echocardiographic cardiac dysfunction, elevated cardiac filling pressure, loop diuretic administration, and NT-pro brain natriuretic peptide > 300 pg/mL. Of 228 dyspnoeic patients, 64 patients (28%) had adjudicated AHF, and 79 patients (35%) had echo-bnp AHF. Patients with AHF were older (78 years vs. 73 years), had lower left ventricular ejection fraction (36% vs. 55%), had higher elevated left ventricular filling pressures (98% vs. 18%), and had higher NT-pro brain natriuretic peptide levels (3628 pg/mL vs. 470 pg/mL). The odds to diagnose adjudicated AHF and echo-bnp AHF were up to four times greater using LDCT (cOR: 3.89 [2.15, 7.06] and cOR: 2.52 [1.45, 4.38], respectively). For each radiologic sign of pulmonary congestion, the LDCT provided superior or equivalent results as the chest radiographs, and the interrater agreement was higher using LDCT (kappa 0.88 [95% CI: 0.81, 0.95] vs. 0.73 [95% CI: 0.63, 0.82]). As first-line imaging modality, LDCT will find one additional adjudicated AHF in 12.5 patients and prevent one false-positive in 20 patients. Similar results were demonstrated for echo-bnp AHF. CONCLUSIONS: In consecutive dyspnoeic patients admitted to the emergency department, LDCT is significantly better than chest radiographs in detecting pulmonary congestion.

Constructing a clinical radiographic knee osteoarthritis database using artificial intelligence tools with limited human labor: A proof of principle.

OBJECTIVE: To create a scalable and feasible retrospective consecutive knee osteoarthritis (OA) radiographic database with limited human labor using commercial and custom-built artificial intelligence (AI) tools. METHODS: We applied four AI tools, two commercially available and two custom-built tools, to analyze 6 years of clinical consecutive knee radiographs from patients aged 35-79 at the University of Copenhagen Hospital, Bispebjerg-Frederiksberg Hospital, Denmark. The tools provided Kellgren-Lawrence (KL) grades, joint space widths, patella osteophyte detection, radiographic view detection, knee joint implant detection, and radiographic marker detection. RESULTS: In total, 25,778 knee radiographs from 8575 patients were included in the database after excluding inapplicable radiographs, and 92.5% of the knees had a complete OA dataset. Using the four AI tools, we saved about 800 hours of radiologist reading time and only manually reviewed 16.0% of the images in the database. CONCLUSIONS: This study shows that clinical knee OA databases can be built using AI with limited human reading time for uniform grading and measurements. The concept is scalable temporally and across geographic regions and could help diversify further OA research by efficiently including radiographic knee OA data from different populations globally. We can prevent data dredging and overfitting OA theories on existing trite cohorts by including various gene pools and continuous expansion of new clinical cohorts. Furthermore, the suggested tools and applied approaches provide an ability to retest previous hypotheses and test new hypotheses on real-life clinical data with current disease prevalence and trends.

Two-year MRI-defined structural damage and patient-reported outcomes following surgery or exercise for meniscal tears in young adults.

OBJECTIVE: To investigate potential differences in structural knee joint damage assessed by MRI and patient-reported outcomes (PROMs) at 2-year follow-up between young adults randomised to early surgery or exercise and education with optional delayed surgery for a meniscal tear. METHODS: A secondary analysis of a multicentre randomised controlled trial including 121 patients (18-40 years) with an MRI-verified meniscal tear. For this study, only patients with 2-year follow-up were included. The main outcomes were the difference in worsening of structural knee damage, assessed by MRI using the Anterior Cruciate Ligament OsteoArthrits Score, and the difference in change in the mean score of four Knee Injury and Osteoarthritis Outcome Score (KOOS4) subscales covering pain, symptoms, function in sport and recreation, and quality of life, from baseline to 2 years. RESULTS: In total, 82/121 (68%) patients completed the 2-year follow-up (39 from the surgical group and 43 from the exercise group). MRI-defined cartilage damage had developed or progressed in seven (9.1%) patients and osteophytes developed in two (2.6%) patients. The worsening of structural damage from baseline to 2-year follow-up was similar between groups. The mean (95% CI) adjusted differences in change in KOOS4 between intervention groups from baseline to 2 years was -1.4 (-9.1, 6.2) points. The mean improvement in KOOS4 was 16.4 (10.4, 22.4) in the surgical group and 21.5 (15.0, 28.0) in the exercise group. No between group differences in improvement were found in the KOOS subscales. CONCLUSIONS: The 2-year worsening of MRI-defined structural damage was limited and similar in young adult patients with a meniscal tear treated with surgery or exercise with optional delayed surgery. Both groups had similar clinically relevant improvements in KOOS4, suggesting the choice of treatment strategy does not impact 2-year structural knee damage or PROMs. TRIAL REGISTRATION NUMBER: NCT02995551.

Current Role of Conventional Radiography of Sacroiliac Joints in Adults and Juveniles with Suspected Axial Spondyloarthritis: Opinion from the ESSR Arthritis and Pediatric Subcommittees.

This opinion article by the European Society of Musculoskeletal Radiology Arthritis and Pediatric Subcommittees discusses the current use of conventional radiography (CR) of the sacroiliac joints in adults and juveniles with suspected axial spondyloarthritis (axSpA). The strengths and limitations of CR compared with magnetic resonance imaging (MRI) and computed tomography (CT) are presented.Based on the current literature and expert opinions, the subcommittees recognize the superior sensitivity of MRI to detect early sacroiliitis. In adults, supplementary pelvic radiography, low-dose CT, or synthetic CT may be needed to evaluate differential diagnoses. CR remains the method of choice to detect structural changes in patients with suspected late-stage axSpA or established disease and in patients with suspected concomitant hip or pubic symphysis involvement. In children, MRI is the imaging modality of choice because it can detect active as well as structural changes and is radiation free.

Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center.

PURPOSE: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. METHODS: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). RESULTS: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %-91 %) and specificity of 90 % (95 % CI: 87 %-92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. CONCLUSIONS: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.

Occurrence and Prediction of Flare After Tapering of Tumor Necrosis Factor Inhibitors in Patients With Axial Spondyloarthritis.

OBJECTIVE: Patients with axial spondyloarthritis (axSpA) in clinical remission tapered tumor necrosis factor inhibitor (TNFi) therapy according to a clinical guideline. Over a 2-year follow-up period, we aimed to investigate flare frequency, dose at which flare occurred, type of flare, and predictors thereof. METHODS: Patients in clinical remission (Bath Ankylosing Spondylitis Disease Activity Index [BASDAI] < 40, physician global score < 40, and without disease activity the previous year) tapered TNFi to two-thirds the standard dose at baseline, half at week 16, one-third at week 32, and discontinued at week 48. Flares were defined as BASDAI flare (BASDAI ≥ 40 and change ≥ 20 since inclusion), and/or clinical flare (development of inflammatory back pain, musculoskeletal or extraarticular manifestations, and/or Ankylosing Spondylitis Disease Activity Score [ASDAS] ≥ 0.9), and/or magnetic resonance imaging (MRI) flare (≥ 2 new or worsened inflammatory lesions). RESULTS: Of 108 patients, 106 (99%) flared before 2-year follow-up: 29 patients (27%) at two-thirds standard dose, 21 (20%) at half dose, 29 (27%) at one-third dose, and 27 (25%) after discontinuation. Regarding type of flare, 105 (99%) had clinical flares, 25 (24%) had BASDAI flares, and 23 (29% of patients with MRI at flare available) had MRI flares. Forty-one patients (41%) fulfilled the Assessment of SpondyloArthritis international Society (ASAS) definition of clinically important worsening (≥ 0.9 increase since baseline). Higher baseline physician global score was an independent predictor of flare after tapering to two-thirds (OR 1.19, 95% CI 1.04-1.41, P = 0.01). Changes in clinical and/or imaging variables in the 16 weeks prior to tapering did not predict flare. CONCLUSION: Almost all (99%) patients with axSpA in clinical remission experienced flare during tapering to discontinuation, but in over half of these patients, flare did not occur before receiving one-third dose or less. Higher physician global score was an independent predictor of flare.

Comparative effectiveness of pharmacological interventions for hand osteoarthritis: a systematic review and network meta-analysis of randomised trials.

OBJECTIVE: To explore the comparative effectiveness of pharmacological interventions for hand osteoarthritis (OA). METHODS: We systematically searched Embase, MEDLINE, and the Cochrane Central Register of Controlled Trials from inception until 26 December 2021, for randomised trials of pharmacological interventions for people with hand OA. Two reviewers independently extracted study data and assessed the risk of bias. We calculated the effect sizes for pain (standardised mean differences) using Bayesian random effects models for network meta-analysis (NMA) and pairwise meta-analysis. Based on a pre-specified protocol, we prospectively registered the study at PROSPERO, CRD42021215393. RESULTS: We included 72 trials with 7609 participants. 65 trials (n=5957) were eligible for the quantitative synthesis, investigating 29 pharmacological interventions. Oral non-steroidal anti-inflammatory drugs (NSAIDs) and oral glucocorticoids' NMA effect sizes were -0.18 (95% credible interval -0.36 to 0.02) and -0.54 (-0.83 to -0.24), respectively, compared with placebo, and the result was consistent when limiting evidence to the pairwise meta-analysis of trials without high risk of bias. Intra-articular hyaluronate, intra-articular glucocorticoids, hydroxychloroquine, and topical NSAIDs' NMA effect sizes were 0.22 (-0.08 to 0.51), 0.25 (0.00 to 0.51), -0.01 (-0.19 to 0.18), and -0.14 (-0.33 to 0.08), respectively, compared with placebo. Oral NSAIDs were inferior to oral glucocorticoids with an NMA effect size of 0.36 (0.01 to 0.72). No intervention was superior to placebo when stratifying for thumb and finger OA. CONCLUSION: Oral NSAIDs and glucocorticoids are apparently effective pharmacological interventions in hand OA. Intra-articular therapies and topical NSAIDs were not superior to placebo.