Nalbant Hande, Abdelhafez Yasser G, Bateni Cyrus, Godinez Felipe, Lee Sonia, Zhang Michelle, Qi Jinyi, Yuan Nimu, Sen Fatma, Moawad Ahmed W, Elsayes Khalid M, Darrow Morgan A, Link Thomas M, Guindani Michele, Nardo Lorenzo
Department of Radiology, University of California Davis, Sacramento, CA, USA.
Radiology and Nuclear Medicine Department, South Egypt Cancer Institute, Assiut University, Assiut, Egypt.
Skeletal Radiol. 2025 Jun 5. doi: 10.1007/s00256-025-04957-8.
To compare the diagnostic performance of non-contrast MRI versus MRI with contrast for differentiating atypical lipomatous tumors (ALT) from lipomas.
This multicenter retrospective study included subjects with a histopathologic diagnosis of lipoma or ALT and a preoperative MRI study with contrast. An experienced musculoskeletal radiologist reviewed the images in two sessions, the first session with non-contrast only, and the second session, including postcontrast sequences. In each session, the radiologist assigned a binary classification (lipoma or ALT) and a 5-point diagnostic score to reflect clinical reporting. Pathology reports served as the reference standard. McNemar's test was used to evaluate the statistical significance of the differences in sensitivity and specificity, while intraclass correlation coefficients (ICC) compared ordinal diagnostic scores.
Four-hundred and forty-one patients (219 ALT, 222 lipoma) were eligible for analysis. In the first session, 76.1% of the lesions were classified as true negative and 76.3% as true positive. In the second session, 74.8% of lesions were assessed as true negative and 77.2% as true positive. There were no significant differences between the two readings in terms of sensitivity or specificity. The agreement in assigning the exact score between the two sessions, as measured by ICC, was 0.878 (95%CI: 0.855-0.898).
Our study found no significant difference between the radiological readings of MRIs that used only precontrast sequences were used for the evaluation of lipoma and ALT, and those that included both pre- and postcontrast-enhanced sequences.
比较非增强MRI与增强MRI在鉴别不典型脂肪瘤（ALT）和脂肪瘤方面的诊断性能。
这项多中心回顾性研究纳入了经组织病理学诊断为脂肪瘤或ALT且术前行增强MRI检查的受试者。一位经验丰富的肌肉骨骼放射科医生分两个阶段对图像进行了评估，第一阶段仅使用非增强图像，第二阶段包括增强后序列。在每个阶段，放射科医生进行二元分类（脂肪瘤或ALT）并给出一个5分的诊断评分以反映临床报告情况。病理报告作为参考标准。采用McNemar检验评估敏感性和特异性差异的统计学意义，同时使用组内相关系数（ICC）比较有序诊断评分。
441例患者（219例ALT，222例脂肪瘤）符合分析条件。在第一阶段，76.1%的病变被分类为真阴性，76.3%为真阳性。在第二阶段，74.8%的病变被评估为真阴性，77.2%为真阳性。两次读片在敏感性或特异性方面无显著差异。通过ICC测量，两个阶段在给出确切评分方面的一致性为0.878（95%CI：0.855 - 0.898）。
我们的研究发现，仅使用对比前序列的MRI读片与同时包括对比前和对比后增强序列的MRI读片在评估脂肪瘤和ALT方面无显著差异。
