Skeletal Radiology最新文献

Objective: To develop and validate an artificial intelligence-based tool for the diagnosis of osteoporosis/osteopenia using hip radiographs. The tool aims to classify femurs into risk-based categories for osteoporosis/osteopenia, enabling patient prioritization, enhancing preventive medicine through incidental detection, and assisting clinicians' diagnosis in general.

Materials and methods: The AI tool was designed to perform three preprocessing tasks before the osteoporosis/osteopenia prediction: (1) splitting images into single femurs, (2) identifying and discarding femurs with prostheses, and (3) cropping images to isolate the proximal femur. A total of 2691 anteroposterior hip radiographs from 1654 patients were included in the study. The osteoporosis/osteopenia prediction model was trained on 3227 single femur images and tested on 826. Additionally, a final evaluation experiment was conducted on 313 new radiographs from 239 patients to assess the tool's applicability.

Results: The tool demonstrated high performance in the preprocessing tasks, achieving 99.0% accuracy in classifying single vs. double femur images, 99.3% accuracy in identifying prosthetic femurs, and 99.2% pixel accuracy in delineating the proximal femur before cropping. The final prediction model achieved an area under the curve of 86.6% for detecting osteoporosis/osteopenia in the test set and 81.0% in the final evaluation experiment.

Conclusions: The obtained results demonstrate the potential of the proposed AI-based pipeline for prediction of osteoporosis/osteopenia using hip radiographs. This study suggests that a tool based on the proposed methods could support DXA triage, incidental osteoporosis detection, and clinical decision-making in settings with limited access to bone densitometry.

Ossifying fibromyxoid tumors (OFMT) are rare tumors with an intermediate risk of malignancy that are usually found in the extremities. To date, only three cases of OFMTs involving the paraspinal musculature have been reported in the literature. Two of these previously reported cases demonstrated poorly circumscribed tumors with involvement of adjacent osseous and soft tissue structures. We report the first case of a well-circumscribed OFMT involving the paraspinal musculature with extrinsic erosion of bone but otherwise no significant soft tissue extension. Specifically, we review the CT and MRI findings, immunohistopathology, and genetics used to confirm the diagnosis of OFMT.

A distal posterior interosseous nerve (DPIN) neuroma related to prior surgery of the wrist can lead to disabling chronic pain. Ultrasound may represent a useful diagnostic tool due to its high resolution and ability to detect iatrogenic neuromas along small nerves of the limbs. However, the utility of ultrasound in the evaluation of both the normal sonographic appearance of the DPIN and traumatic neuromas is limited. We present a series of two patients who suffered from chronic dorsal wrist postoperative pain, where ultrasound detected histologically confirmed DPIN neuromas. After DPIN resection, the pain of our two patients completely disappeared without any functional deficit.

Objective: To introduce a tear location classification system for posterior cruciate ligament (PCL) injuries, aiming to unify diagnostic criteria and improve clinical decision-making based on tear location.

Materials and methods: A retrospective analysis was conducted using magnetic resonance imaging (MRI) examinations from a single institution, identifying all patients with acute or subacute PCL injuries from 2008 to 2024. Ligament sprains without significant fiber disruption and chronic injuries were excluded. Tears were classified by four independent observers according to the relative length (%) of the intact distal remnant compared to the total PCL length: Type I (> 90%), Type Ib (femoral bony avulsion), Type II (90-75%), Type III (75-25%), Type IV (distal tear 10-25%), Type V (< 10%), and Type Vb (tibial bony avulsion). For intra-observer reliability analysis, measurements were repeated once by two observers after 4 weeks.

Results: A subset of 45 MRIs with diagnosed PCL injuries that met the inclusion criteria (mean age 40.5 ± 19.2 years, 40% female) had 24% proximal tears, 33% midsubstance tears, 33% distal tears, and 11% single-bundle injuries. Inter-observer reliability (Fleiss' kappa, 0.88; 95% CI, 0.84-0.94; p < 0.01) and intra-observer reliability demonstrated to be almost perfect (Cohen's kappa, 0.95; 95% CI, 0.86-0.99 & 0.92; 95% CI, 0.84-0.99).

Conclusion: The proposed MRI-based classification system-which includes femoral avulsion, proximal quarter, midsubstance (50%), distal quarter, and tibial avulsion injuries-offers a reliable method for identifying the anatomical location of partial and complete PCL injuries. Standardizing tear localization has the potential to improve diagnostic consistency and inform more tailored, evidence-based treatment strategies.

Level of evidence: Diagnostic study; III.

An untreated scapholunate interosseous ligament injury or scaphoid nonunion leads to progressive post-traumatic periscaphoid osteoarthritis and carpal collapse. Symptomatic forms, known as scapholunate advanced collapse (SLAC) or scaphoid nonunion advanced collapse (SNAC), are common and disabling, requiring surgical management. Numerous operative options exist depending on the stage of collapse and the extent of joint involvement. We apply a modified Watson radiological classification consisting of four stages of pancarpal collapse, integrating the etiological origin (SLAC or SNAC) and the arthritic progression pattern. This review summarizes current surgical principles for each stage of the disease and highlights the key role of imaging in postoperative assessment. Radiography remains the first-line tool to evaluate carpal alignment and implant positioning. CT and CBCT provide high-resolution analysis of bony structures and fixation devices, while ultrasound enables dynamic evaluation of superficial soft tissues and implant-tendon relationships without metallic artifacts. MRI, despite its sensitivity for soft tissues and bone marrow, remains limited by metal-induced distortion. A comprehensive understanding of surgical techniques, biomechanical principles, and implant types is essential to accurately interpret postoperative outcomes and detect complications early.

Objective: To evaluate multiparametric MRI features of pediatric soft-tissue sarcomas, comparing pre-treatment and post-treatment features, and assessing correlation with clinical outcomes.

Materials and methods: Retrospective cohort study, including pediatric patients (≤ 18 years) with histologically-confirmed soft-tissue sarcomas who underwent MRI with anatomic and functional sequences in consecutive series. Post-treatment MRI was available for a subset, and features were recorded by two readers. Pre-treatment and post-treatment features were compared using Wilcoxon signed-rank test with Hodges-Lehmann estimator. Inter-reader agreement was assessed with intraclass correlation coefficient, and Mann-Whitney and Cox regression tests were used to correlate the features with clinical outcome.

Results: There were 27 patients. Pre-treatment we measured tumor size, ADCminimum (mean: 507 × 10-6mm²/s) and ADCaverage (mean: 690 × 10-6mm²/s). Post-treatment (10 patients) included a reduction in tumor size (p = 0.002) and higher ADCminimum (mean: 1116 × 10-6mm²/s) and ADCaverage (mean: 1344 × 10-6mm²/s) values. Pre-treatment size was larger in patients with metastasis at baseline (p = 0.004) and progression (p = 0.002), but size change after treatment did not correlate with progression. Baseline ADC​​ did not correlate with progression, but the group with progression showed less difference between pre- and post-treatment ADCminimum (p = 0.019) and ADCaverage (p = 0.032). There was excellent agreement between the readers measuring ADCminimum (ICC = 0.991) and ADCaverage (ICC = 0.978).

Conclusion: For high-grade pediatric soft-tissue sarcomas, the pre-treatment size is an important prognostic factor, and a reduction in tumor size was observed after treatment, but did not correlate with progression. However, changes to ADC values between pre-treatment and post-treatment MRI correlated with disease progression, suggesting that ADC is a potentially useful biomarker of clinical outcome.

Objective: To retrospectively evaluate outcomes of an ultrasound-guided A1 pulley fenestration release technique using small-gauge (hypodermic or spinal) needles for the treatment of trigger finger (TF).

Materials and methods: A retrospective chart review of all TF fenestration release procedures performed by two musculoskeletal radiologists between July 2020 and August 2024 was conducted. The technique included a steroid injection after release. Preprocedural and postprocedural functional Quinnell grades and any immediate complications from the procedure report were primary outcome measures. Clinical pain score, functional outcomes, other TF interventions, and delayed complications were secondary outcome measures.

Results: A total of 119 procedures were performed in 92 patients (61% women, mean age 63 ± 13 years), with 95 procedures (80%) following prior TF injection with refractory symptoms. Periprocedural Quinnell grades were reported in 99 procedures (83%), with immediately improved scores for all (median-preprocedural-to-postprocedural, 3-to-0; p < 0.001). Retrospective follow-up data were available for 60 procedures (50%), of which 70% experienced functional improvement. Pain scores were significantly improved at follow-up (median-preprocedural-to-follow-up, 4-to-0.5; p = 0.046). Subsequent interventions occurred following 22 cases (18%), of which 15 (13%) required additional steroid injections, 3 (3%) required repeat fenestration, and 5 (4%) required surgical release. No immediate or delayed complications were otherwise reported. Higher Quinnell grade at end of procedure predicted increased rates of follow-up (OR = 3.17, p = 0.012) and suggested worse functional status at follow-up (OR = 0.25, p = 0.054), and smaller peri-procedural improvement increased odds of additional intervention (OR = 0.48, p < 0.001).

Conclusion: Ultrasound-guided fenestration is an effective and safe treatment for TF in the outpatient setting.

Objective: The goal of this study is to compare the sensitivity, accuracy, and specificity of detection of musculoskeletal sarcoma local recurrence (LR) using axial-only MRI compared with multiplanar MRI.

Materials and methods: This retrospective study included 101 patients (47 with pathology-proven LR, 54 without recurrence) who underwent pre- and post-operative MRIs between 2013 and 2023 at a single academic center. Three fellowship-trained musculoskeletal radiologists independently reviewed anonymized MRIs. Initial review included only the axial sequences from the surveillance MRIs. Secondary review following a 6-month washout included all sequences from the multiplanar surveillance MRIs.

Results: No statistically significant differences in accuracy, sensitivity, and specificity were observed between axial-only and complete (multiplanar) MRI analyses (p > 0.05). Accuracy of LR detection with axial-only vs. complete (multiplanar) MRI was 94.5% vs. 93.1% (Reader 1), 87.1% vs. 91.1% (Reader 2), and 82.7% vs. 86.1% (Reader 3). LR size ranged from 0.7 to 8.9 cm (mean 3.2 cm, median 2.6 cm). All missed recurrences when using axial-only surveillance MRI measured < 2 cm. Inter-reader agreement improved from 0.768 (axial-only) to 0.834 (complete MRI).

Conclusion: Surveillance MRI using axial sequences alone demonstrated comparable accuracy, sensitivity, and specificity to multiplanar MRI for detecting musculoskeletal sarcoma LR.