Radiologie (Heidelberg, Germany)最新文献

Background: Intramedullary spinal cord tumors are rare neoplasms but frequently cause significant neurological deficits because of the limited compensatory capacity of the spinal cord. Magnetic resonance imaging (MRI) is the imaging modality of choice and often allows a reliable differential diagnosis based on characteristic imaging findings.

Objectives: To provide a practical neuroradiological approach to the diagnosis of intramedullary spinal cord tumors, including recent developments in the World Health Organization (WHO) classification.

Materials and methods: Narrative review of the current literature regarding epidemiology, imaging characteristics, WHO classification, and differential diagnosis of intramedullary spinal cord tumors.

Results: Ependymomas, astrocytomas, and hemangioblastomas account for the majority of intramedullary tumors. Tumor location, longitudinal extent, enhancement pattern, hemorrhagic components, syringomyelia, and flow voids substantially narrow the differential diagnosis. Recent WHO classifications increasingly incorporate molecular markers, including MYCN-amplified spinal ependymoma and H3 K27-altered diffuse midline glioma.

Conclusion: Structured MRI assessment frequently enables a reliable preoperative diagnosis. Inflammatory, vascular, and ischemic disorders must always be considered as important differential diagnoses.

Background: Primary bone tumors of the spine are rare causes of spinal symptoms and often present with nonspecific clinical findings. Reliable differentiation from degenerative or inflammatory conditions is usually only possible by imaging. Precise anatomical compartment allocation as well as the combination of magnetic resonance imaging (MRI) and computed tomography (CT) are essential for diagnosis.

Objective: The aim of this review is to provide an overview of primary benign and malignant bone tumors of the spine, focusing on their imaging characteristics and differential diagnostic features.

Materials and methods: A narrative review of typical spinal bone tumors, considering clinical presentation, pathophysiology, and radiologic findings. Benign lesions included osteoma, osteoid osteoma, osteoblastoma, osteochondroma, vertebral hemangioma, and aneurysmal bone cyst, while malignant entities comprised chondrosarcoma and chordoma.

Results: Benign tumors demonstrate characteristic imaging patterns: osteomas appear as homogeneous sclerotic lesions without relevant contrast enhancement, whereas osteoid osteomas typically present with a small vascularized nidus and surrounding sclerosis. Osteoblastomas are usually larger and expansile-destructive. Osteochondromas show continuity of the cortex and medullary cavity with the host bone. Vertebral hemangiomas demonstrate characteristic trabecular patterns, while aneurysmal bone cysts typically show fluid-fluid levels on MRI. Malignant tumors such as chondrosarcomas present as osteolytic lesions with chondroid matrix calcifications, whereas chordomas appear as locally destructive tumors with marked T2 hyperintensity and soft-tissue extension.

Conclusion: Imaging plays a central role in the diagnosis and differential diagnosis of primary spinal bone tumors. MRI is particularly important for assessing tumor extent and soft-tissue involvement, whereas CT enables detailed evaluation of matrix mineralization and cortical destruction. Despite characteristic imaging features, overlap between benign and malignant lesions exists, making interdisciplinary radiologic-pathologic correlation frequently necessary.

Background: Bone marrow reconversion is a physiological response to increased hematopoietic demand and is common in chronic anemia. Magnetic resonance imaging (MRI) enables noninvasive assessment of marrow composition; however, visual evaluation remains subjective. Because the psoas signal is largely unaffected by hematologic status, it can serve as a stable internal reference.

Objectives: To quantify lumbar vertebral marrow signal alterations in women with iron deficiency anemia (IDA) and evaluate the diagnostic performance of the T1-weighted signal ratio between the lumbar vertebrae and the psoas muscle as a predictor of anemia.

Materials and methods: This retrospective study analyzed lumbar spine MRIs and laboratory data of 141 women (83 IDA, 58 controls). On sagittal T1-weighted images, circular regions of interest were placed in the L1 vertebral body and adjacent psoas muscle to calculate the vertebra/psoas signal ratio. Group comparisons, correlation, and receiver operating characteristic (ROC) analysis with 95% confidence intervals were performed.

Results: The mean vertebra/psoas ratio was significantly lower in IDA (1.63 ± 0.34) than in controls (2.19 ± 0.71; p < 0.001). Hemoglobin levels correlated positively with the ratio (β = 0.11; R² = 0.16; p < 0.05). The ROC analysis results demonstrated good discrimination (AUC = 0.81; 95% CI 0.73-0.88); an optimal cut-off of 1.93 yielded 85.5% sensitivity and 63.8% specificity.

Conclusion: The lumbar vertebra/psoas T1-weighted signal ratio is significantly reduced in IDA and correlates with hemoglobin levels. It represents a simple and reproducible quantitative MRI biomarker that may aid in the screening and diagnostic evaluation of anemia on routine spine imaging.

Objective: This study aimed to explore the anatomy and injury patterns of trapeziometacarpal (TMC) joint ligaments by using magnetic resonance imaging (MRI).

Materials and methods: In total, 32 asymptomatic healthy volunteers (16 women, 16 men; mean age, 34.5 ± 5.4 years) and 18 patients (12 women, 6 men; mean age, 43.6 ± 10.4 years) were recruited between January 2019 and November 2024. All participants underwent MRI of the TMC joint using T1-weighted imaging and proton density-weighted imaging with fat suppression. Two musculoskeletal radiologists independently analyzed the MRI features of the TMC joint ligaments in volunteers and patients.

Results: In asymptomatic volunteers, the normal anterior oblique ligament (AOL), dorsal radial ligament (DRL), and posterior oblique ligament appeared as slender, low-signal fibrous structures on MRI, whereas the normal intermetacarpal ligament exhibited a striated signal. The AOL was significantly thinner in women than in men, and there were no significant sex differences for other ligaments. Ligament injuries were characterized by changes in MRI signal: Sprains showed high-signal alterations, partial tears displayed thickening and partially discontinuous fibers with disrupted signals, and complete ruptures were characterized by ligament retraction and fluid-filled gaps. Among the ligaments, the AOL was most susceptible to injury, followed by the DRL.

Conclusion: Magnetic resonance imaging is an accurate method for assessing the anatomy and pathological conditions of the TMC joint ligaments, thus aiding in the accurate diagnosis and management of ligament injuries.

Background: Osteoporosis is frequently underdiagnosed in trauma patients. However, contrast-enhanced whole-body computed tomography (WBCT) in polytrauma care enables opportunistic assessment of bone status without additional radiation exposure. The aim of this study was to investigate whether vertebral, femoral, and dentomaxillofacial CT parameters can identify patients with documented osteopenia or osteoporosis and whether a combined score improves classification.

Materials and methods: This retrospective, single-center study included 60 WBCT cases. Twenty patients with osteopenia or osteoporosis documented in their initial trauma report were compared with 20 age- and sex-matched reference patients and 20 younger patients. The final classification of bone status was based on an evaluation of patient records, including dual-energy X‑ray absorptiometry (DXA) data (where available), and medical history. Trabecular density was measured in the lumbar spine (L1-L3), femoral neck, and maxillary alveolar ridge. Hard palate cortical thickness (HPT) was measured in millimeters. The whole-body osteoporosis screening score was defined as WBOS = ƒ(HPT, hip HU, lumbar HU).

Results: Measurements of HPT, maxilla, hip, and lumbar spine decreased significantly in patients with osteopenia/osteoporosis patients. The combined WBOS score demonstrated the best exploratory performance, with a sensitivity of 100.0%, a specificity of 97.1%, an accuracy of 98.3%, and an AUC of 0.994 at a preliminary threshold of 361.

Conclusion: Opportunistic multisite assessment of bone status using WBCT in polytrauma appears feasible. The WBOS showed the best exploratory discrimination but requires further validation against DXA or quantitative CT before clinical implementation.

Background: We aimed to explore the use of deep learning image reconstruction (DLIR) to improve virtual non-contrast (VNC) images from enhanced dual-energy computed tomography (DECT).

Materials and methods: A total of 91 patients undergoing treatment for liver lesions from September to December 2024 were analyzed. All patients underwent true non-contrast (TNC) and dual-phase enhanced DECT with medium-level DLIR (DLIR-M). The VNC images were generated from arterial phase (VNC_A) and venous phase (VNC_V) scans. The CT values and standard deviations of liver, spleen, bilateral erector spinae, and liver lesions were measured across the three image groups, and the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. Two radiologists subjectively scored the image quality. Measurements and quality scores were statistically compared among the groups.

Results: There was no difference in CT values among the three groups (P > 0.05); VNC_A and VNC_V images did not alter the primary imaging characteristics of hypervascular liver lesions. The VNC_A and VNC_V images had lower image noise than the TNC images, with VNC_V images having the lowest noise. Except for the spleen, VNC_A and VNC_V images had higher SNR than TNC images, with VNC_V images having the best SNR. The subjective scores of the TNC, VNC_A, and VNC_V images were all greater than 3, showing good consistency between the two observers (kappa = 0.78, 0.76, and 0.83, respectively).There was no difference between the subjective evaluation of VNC_A and TNC (P > 0.05), while VNC_V had better image quality than TNC (P < 0.05). The total effective radiation dose was 8.55 ± 0.38 mSv; thus, replacing TNC with VNC would reduce the dose by 33.4%.

Conclusion: Virtual non-contrast images generated in liver DECT with DLIR provide similar CT values to TNC images. The VNC_V images provide lower image noise as well as higher SNR and subjective scores than TNC and are recommended for clinical use.

Advances in knowledge: Virtual non-contrast images with DLIR generated from DECT can improve image quality and reduce radiation dose.

Guidelines and directives are instruments with varying legal effects that are relevant to liability and cost law. While guidelines primarily gain legal relevance in the question of how standard-compliant diagnostics and therapy can be expected under a physician contract, directives, with their cost sensitivity (which does not apply in liability law!), play a central role in the question of benefits and reimbursement in the statutory health insurance system. In borderline areas, it should be noted that both instruments can also become significant in the other sector.

Background: Teleradiology has evolved into an integral component of radiological care. Among approximately 1900 hospitals in Germany, only around 650 maintain their own radiological departments; patient care is increasingly provided through cooperative arrangements incorporating teleradiological expertise. However, the spatial separation between image acquisition and diagnostic interpretation presents specific liability-related challenges.

Objective: This article examines the central liability risks in teleradiology and provides concrete recommendations for everyday radiological practice. The aim is to protect physicians from legal consequences and to guarantee the quality of patient care.

Results: The legal foundations are based on §§ 630a et seq. of the German Civil Code (BGB). Teleradiologists are subject to the same specialist standard as colleagues working on-site. Specific risks include technical transmission problems, time-critical emergencies, and organizational liability issues. Recommendations include adherence to technical standards (Digital Imaging and Communications in Medicine, DICOM), virtual private network (VPN) encryption, quality assurance prior to interpretation, redundant data transmission pathways, written cooperation agreements, and emergency protocols. Professional liability insurance must explicitly cover teleradiological activities.

Conclusion: Legal jurisprudence has not yet provided definitive answers, particularly regarding technical transmission errors. Quality assurance, documentation, and clear allocation of responsibilities are central to avoiding liability. For cross-border activities, the territorial scope of insurance coverage must be verified. Furthermore, regular review of insurance coverage is recommended.

Background: Artificial intelligence (AI) is increasingly used in radiological diagnostics, particularly for screening, detection, and prioritization of imaging exams. While these systems promise efficiency gains and improved diagnostic accuracy, their clinical integration raises important legal questions concerning responsibility and liability in cases of diagnostic error.

Objectives: This article analyzes the legal framework governing AI in radiology and examines the allocation of liability between physicians, health care institutions, and manufacturers within existing medical liability structures.

Materials and methods: A doctrinal legal analysis was conducted based on European and German legal sources, including the AI Act and the Medical Device Regulation, as well as provisions of German medical liability law.

Results and conclusion: AI systems in radiology are legally classified as medical tools rather than independent actors. Consequently, they cannot bear legal responsibility themselves. Liability primarily remains with the physician, who must critically review and validate AI-generated outputs within the diagnostic process. Manufacturers may be liable under product liability and regulatory compliance frameworks. The interaction between technological complexity and existing evidentiary standards creates increasing tension within traditional liability doctrines. Despite technological progress, the final diagnostic decision and legal responsibility remain with the physician. AI currently functions as a decision-support tool rather than a substitute for medical judgment. However, the increasing relevance of autonomous systems suggests that a distinct legal framework for AI in medicine may emerge in the future.