Purpose: The aim of this retrospective study was to report the prevalence and imaging characteristics of bone metastases detected with F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) and, when possible, compare these findings with the performance of bone scans in the same patients.
Methods: The reports of 403 patients with histologically proved malignant disease who underwent a PET scan for initial or post-therapeutic staging were reviewed for the presence of possible bone metastases. Based on the final diagnosis confirmed by histopathologic analysis or clinical follow-up, the PET findings of patients with positive bone metastases were evaluated in terms of location, intensity, and patterns. When the PET scan was positive, the PET results were compared with the findings of available bone scans.
Results: PET studies suggested the presence of bone metastases in 38 patients (9%). No follow-up data were available for 9 patients, and the remaining 29 were evaluated further. Of these patients, 6 had false-positive findings, whereas bone metastatic involvement was clinically confirmed in 23 patients. The primary malignant findings included lung cancer (n = 9), esophageal cancer (n = 3), lymphoma (n = 2), melanoma (n = 2), thyroid cancer (n = 2), breast cancer (n = 1), colon cancer (n = 1), prostate cancer (n = 1), testicular cancer (n = 1), and nasopharyngeal cancer (n = 1). On PET, 5 patients had a solitary metastatic focus (22%), and the remaining 18 patients had multiple lesions (78%). The vertebrae were the most frequently involved bones (74%), followed by pelvic bones (70%), ribs (65%), upper extremities including the scapula (48%), sternum (43%), and lower extremities (43%). The patterns of abnormal uptake were classified into three groups: focal (15 patients, 65%), diffuse (2 patients, 9%), and a mixed pattern (6 patients, 26%). Most of the lesions showed intense abnormal uptake (18 patients, 78%); 5 patients had both intense and moderate FDG uptake. Thirteen of the 23 patients with confirmed bone metastases also had a bone scan, which revealed positive bone disease in all of these patients. However, PET consistently revealed more metastatic foci than did the bone scan on a lesion basis.
Conclusions: The most frequent pattern of detectable bone metastases with FDG-PET imaging was multiple foci of intense uptake. PET revealed more lesions than did bone scanning, independent of the type of cancer or location of bone involvement, in patients who were accurately diagnosed by FDG-PET imaging.