The Performances of Radiographic Criteria for Bone Malignancy When Applied to Computed Tomography and Magnetic Resonance Imaging
2017
Abstract Background The conventional radiologic features that differentiate benign from malignant bone lesions were originally described using radiography (x-ray [XR]). When evaluating sectional imaging studies such as magnetic resonance imaging (MRI) and computed tomography (CT), one may apply these principles to identify malignant bone lesions. The aim of this study was to evaluate the performances of these radiographic features for detecting malignity when applied to CT and MRI. Materials and Methods This retrospective study was approved by our institutional ethical board. Thirty-nine patients with histopathologic proof of a high-grade bone malignancy and preoperative imaging data obtained with a minimum of two different modalities were included in the study. Four radiologists reviewed the images and scored the lesions for distinctness of margins, presence and type of periosteal reaction, matrix mineralization, and presence of soft tissue mass. The average score for each modality was then tested for accuracy with regard to the histopathology. Results When lesion margins were considered, XR was the best modality to detect a high-grade malignancy. MRI, especially postcontrast T1-weighted sequence, was the least helpful in this regard. There was no significant difference between CT and XR and between CT and MRI. When the periosteal reaction was considered, XR was the best modality to detect the malignant type of periosteal reaction. In this regard, MRI and CT were misleading; either by not detecting or undergrading periosteal reaction. MRI was the best modality to detect soft tissue mass. Conclusion Conventional imaging criteria for bone malignancy can be misleading when applied to MRI or CT. When cross-sectional imaging features contradict those from XR, the latter should be the guide for clinical management.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
9
References
3
Citations
NaN
KQI