Objective: To evaluate the efficacy of F-18-fluoro-2-deoxy-D: -glucose positron emission tomography/computed tomography (FDG PET/CT) in differentiating malignant from benign pathologic fractures.
Materials and methods: F-18 FDG PET/CT was performed on 34 patients with pathologic fractures between May 2004 and June 2007. Fractures were located in tubular bones (26), in the pelvis (six), in the spine (one) and in a rib (one). The FDG uptake pattern at the fracture site was described, whether FDG uptake occurred in the marrow or cortex and soft tissue. Maximum standardized uptake values (SUVmax, the largest value at the region of interest) were measured at the fracture site, including cortical bone, bone marrow and soft tissue. As a reference standard, biopsy was used for 12 patients and clinical follow-up for 22 patients. Sensitivity, specificity and diagnostic accuracy of PET/CT were calculated.
Results: There were 19 malignant and 15 benign fractures. In the malignant fractures, PET/CT demonstrated high (mean SUVmax 12.0, range 4.3 to 45.7) F-18 FDG uptake in bone marrow in most cases (17 of 19). In benign fractures, there was low FDG uptake (mean SUVmax 2.9, range 0.6 to 5.5) within cortical bone or adjacent soft tissue around the fracture, rarely in the marrow. There were significant differences in the pattern of intramedullary FDG uptake (P < 0.001) and in the mean SUVmax (P < 0.01) between malignant and benign fractures. The sensitivity, specificity and diagnostic accuracy of F-18 FDG PET/CT were 89.5%, 86.7% and 88.2%, respectively, with a cut-off SUVmax set at 4.7. The time interval between fracture and PET/CT did not significantly influence FDG uptake at the fracture site.
Conclusion: F-18 FDG PET/CT reliably differentiated between malignant and benign fractures based on the SUVmax and based on medullary uptake, which was characteristic for malignant fractures.