Torsten Diekhoff

Background: To compare the performances of magnetic resonance imaging (MRI) and radiography in detecting changes in structural lesions of the sacroiliac joints (SIJs) in routine care patients with axial spondyloarthritis (axSpA). Methods: SIJ MRIs and radiographs of routine care patients with axSpA in four European countries were collected as part of the EuroSpA Imaging Project. Patients were included if both MRIs and radiographs (≤ 6 months apart) were available at two time points (≥ 1 year apart). Images were evaluated centrally and independently by two trained readers (blinded to chronology and other imaging), applying the Spondyloarthritis Research Consortium of Canada SIJ scores (MRI) and the modified New York criteria (mNY, radiography). Intermodality agreement was assessed cross-sectionally and for change. Results: One-hundred-and-five patients (65 male, mean (SD) age 38.1 (10.9) years, symptom duration 9.1 (9.7) years), with a follow-up interval of 2.6 (1.4) years) were included. Baseline agreement between radiography and MRI was slight-to-fair for erosion detection (kappa = 0.12 to 0.42), poor to slight for sclerosis (− 0.04 to 0.07), and substantial to almost perfect for ankylosis (0.66–0.86). Agreement for change over time was poor-to-slight (− 0.04 to 0.20). While lesions remained largely stable on radiography (≤ 6 patients showing change in scores, per lesion type), MRI detected change in at least one lesion in ≥ 50% of patients (positive/negative change in scores (“progression”/”regression”): sclerosis 27/30, erosion 51/38, ankylosis 14/5). A composite MRI score of damage correlated well with mNY grades (Spearman rho = 0.65–0.69). Conclusions: Compared to radiography, MRI detected more SIJ structural lesion changes over time, reflecting its ability to capture dynamic changes not readily visible on radiography.

Objective: To evaluate the accuracy of a commercial deep learning algorithm in measuring spinopelvic parameters on full spine radiographs. Materials and methods: This retrospective study analyzed total spine X-rays from a clinical cohort, assessing spinopelvic parameters (coronal Cobb angle, thoracic kyphosis, lumbar lordosis, pelvic incidence, sagittal vertical axis). Measurements were performed by clinical radiologists, a trained research reader serving as ground truth, and a commercial AI software. Inter-rater reliability was determined via intraclass correlation coefficients between an orthopedic surgeon and the trained reader ( n = 50 images). Performance of both the AI and radiologists was statistically compared to the ground truth using mean absolute error, intraclass correlation coefficients, the Spearman correlation, and diagnostic accuracy metrics (sensitivity, specificity, predictive values). Results: Four hundred ninety-five radiographs were analyzed. Inter-rater reliability between human raters, assessed via intraclass correlation coefficients, was excellent (0.94–1). Agreement between the algorithm and ground truth, also assessed via intraclass correlation coefficients, was good to excellent (0.79–0.99), except for lumbar lordosis (0.68, moderate). Mean absolute errors were lowest for coronal Cobb angles (4,6°; 95% confidence interval = 3.6–5.5°) and highest for lumbar lordosis (8.1°; 95% confidence interval = 6.9–9.3°). The Spearman rank correlations ranged from 0.74 to 0.99, and sensitivity was moderate to excellent (72.6–94.8). These results closely matched those obtained when comparing radiologists to ground truth. In a subgroup of patients with prior spinal fusion surgery, the correlation between the algorithm predictions and the ground truth was reduced, and measurement deviations were higher compared to non-instrumented patients. Conclusion: The commercial AI software predicted most spinopelvic parameters with good reliability and accuracy, coinciding with radiologists in clinical practice, however, showed limitations in patients with spinal instrumentation.

The purpose of this study was to evaluate the effectiveness of collagen-sensitive dual-energy computed tomography (DECT) as a quantitative imaging tool for the assessment and monitoring of load-induced tendinopathy in the Achilles and patellar tendons, comparing it to magnetic resonance imaging (MRI). In a prospective study, 15 consecutive patients clinically diagnosed with Achilles or patellar tendinopathy underwent bilateral DECT and MRI at baseline and 6 months. Quantitative measurements included collagen density assessed via DECT and signal intensity via MRI. Clinical symptoms were evaluated using numerical pain ratings and VISA-A/P scores. The diagnostic accuracy of both imaging modalities was assessed using ROC analysis, and correlations between DECT and MRI findings were investigated. DECT revealed significantly lower collagen densities on corresponding maps in affected tendons (n = 18, 23.7 ± 20.2) compared to unaffected tendons (n = 12, 60.2 ± 29.6 HU, p < 0.001), whereas MRI demonstrated increased signal intensities in pathological regions. ROC analysis indicated comparable diagnostic performance for DECT (AUC = 0.84) and MRI (AUC = 0.80). A strong inverse correlation (r = -0.83) was observed between DECT-measured collagen densities and MRI signal intensities. Clinical improvements at follow-up were reflected by normalization trends in both imaging modalities, though not statistically significant. Collagen-sensitive DECT provides a reliable quantitative approach for detecting and assessing tendon pathologies in load-induced tendinopathy, demonstrating diagnostic capabilities comparable to MRI while offering the possibility for collagen density quantification.

Crystal arthropathies of the hand encompass a wide spectrum of disorders characterized by the deposition of various crystals within joints and soft tissues, often leading to inflammatory and degenerative changes. Due to their ability to mimic other rheumatic diseases, accurate diagnosis is challenging and relies heavily on imaging. This review highlights the pathophysiologic features and typical manifestations of key crystal arthropathies—calcium pyrophosphate crystal deposition, gout, hydroxyapatite deposition disease, oxalosis, amyloidosis, and rare mimickers such as ochronosis—focusing on their radiologic appearances. Emphasis is placed on the strengths and limitations of different imaging modalities, particularly radiography, ultrasound, computed tomography, magnetic resonance imaging, and dual-energy computed tomography, in detecting both crystal deposits and associated inflammatory activity. Understanding these imaging patterns is essential for an appropriate differential diagnosis and clinical management.

Purpose: This study aimed to evaluate the radiological and patient‐reported outcome measures (PROMs) following spheroid‐based matrix‐induced autologous chondrocyte implantation combined with autologous bone grafting (MABCI) for the treatment of osteochondral defects of the knee. Methods: In this retrospective cohort analysis of prospectively collected data, 28 consecutive patients with 29 defects of the femoral condyle (male: 15, female: 13, mean age 27.3 ± 9.8 years, mean defect size 3.7 ± 1.6 cm²) treated with MABCI were included. Radiological assessments were performed using 3T magnetic resonance imaging (MRI) to evaluate cartilage thickness, T2 relaxation times and Magnetic Resonance Observation of Cartilage Repair Tissue (MOCART) 2.0 scores. Clinical outcomes were assessed using validated PROMs, including the International Knee Documentation Committee (IKDC) score, the Knee Injury and Osteoarthritis Outcome Score (KOOS), the Lysholm scale and the Patient‐Reported Outcomes Measurement Information System 29‐Item Profile (PROMIS‐29). Correlation analyses were conducted to evaluate associations between radiological parameters and PROMs. Results: The analysis was conducted at a mean follow‐up of 44.4 ± 21.6 months. Cartilage thickness in the repair tissue exceeded that of the reference cartilage (3.4 ± 0.9 mm vs. 3.0 ± 0.5 mm; p = 0.036), while T2 relaxation times in the repair tissue were within physiological range (50.5 ± 40.0 ms vs. 41.3 ± 5.4 ms; n.s.). The mean MOCART 2.0 score was 61.3 ± 13.0 (volume fill: 16.1 ± 3.7, integration with adjacent cartilage: 12.9 ± 3.3), reflecting satisfactory repair quality. All clinical scores showed significant improvement compared to the preoperative condition (IKDC: 70.6 ± 13.2 vs. 54.8 ± 17.5, p < 0.001; KOOS: pain 78.7 ± 16.9 vs. 66.1 ± 18.6, p = 0.002, symptoms 75.6 ± 16.0 vs. 65.8 ± 19.1, p = 0.006, ADL 86.8 ± 14.0 vs. 77.6 ± 16.8, p = 0.003, sport 63.0 ± 24.2 vs. 47.3 ± 26.9, p = 0.009, quality of life (QoL) 51.0 ± 22.7 vs. 39.6 ± 17.5, p = 0.014; numeric pain rating scale: 3.0 ± 2.2 vs. 5.2 ± 2.4, p < 0.001). A greater cartilage thickness was associated with reduced pain intensity ( r = −0.457, p = 0.013) and improved QoL ( r = 0.429, p = 0.020), indicating that better structural repair is associated with improved patient satisfaction. Conclusions: MABCI results in satisfactory cartilage regeneration and subchondral remodelling in addition to significant improvements in patient satisfaction and overall QoL. Level of Evidence: Level III.

Zusammenfassung: Die Dual-Energy-Computertomografie (DECT) hat sich in den letzten Jahren als wertvolles Werkzeug in der rheumatologischen Bildgebung etabliert. Neben der Detektion von Harnsäurekristallen bei Gicht erlaubt sie durch virtuelle Kalziumsubtraktion auch die Darstellung von Knochenmarködemen, nach Kontrastmittelgabe die Detektion aktiver Weichteilentzündung. Gerade bei komplexen Differenzialdiagnosen oder Patienten mit MRT-Kontraindikationen bietet DECT einen erheblichen Zusatznutzen. Der Beitrag beleuchtet technische Hintergründe, Limitationen und klinische Einsatzmöglichkeiten in der muskuloskelettalen Diagnostik, insbesondere bei Gicht und rheumatoider Arthritis. Auch neue Entwicklungen wie photonenzählende CT-Detektoren und multiparametrische Auswertungen werden vorgestellt.