Primary bone lymphoma (PBL) is one of the rarest primary bone malignancies. Very limited data are available on the role of fluorine-18-fluorodeoxyglucose (F-FDG) PET/CT in PBL. Overall, 16 relevant patients had F-FDG PET/CT for staging or staging/restaging PBL. The cases with extraskeletal lesions such as lymphadenopathy or with osseous involvements in disseminated lymphoma were excluded on the basis of the definition of PBL. F-FDG PET/CT findings were correlated to pathologic and/or anatomic image information. F-FDG PET/CT was positive with markedly increased uptake in all known PBL lesions, with a sensitivity of 100%. In 15 of 16 (93.8%) cases, the bone lesions infiltrated the adjacent soft tissue with visible soft tissue components. Except for known primary bone lesion, F-FDG PET/CT showed additional 24 bone lesions in seven patients, which accounted for 44% of all cases. With maximum standardized uptake value of 2.5 as a cutoff for differentiating residual lymphoma from metabolically inactive disease on restaging scan, the sensitivity, specificity, positive predictive value, and negative predictive value of F-FDG PET/CT for PBL were 100% (4/4), 75% (9/12), 57% (4/7), and 100% (9/9), respectively. The diagnostic accuracy of restaging F-FDG PET/CT was 81% (13/16). Two of three false-positive scans were because of post-therapeutic osteonecrosis. F-FDG PET/CT is sensitive for showing PBL that is typically highly F-FDG avid and has the ability to depict extraskeletal soft tissue involvement and additional bone lesions on staging. More importantly, F-FDG PET/CT may accurately and reliably evaluate therapeutic response with excellent sensitivity and negative predictive value. However, false-positive F-FDG uptake poses a concern on restaging PET/CT. Post-therapeutic osteonecrosis is a potential source of false-positive findings on F-FDG PET/CT.