Pattern-Based Interpretation Criteria for 18F-Fludeoxyglucose Positron Emission Tomography/Computed Tomography in the Assessment of Pyogenic Spine Infection

摘要： 18F-?udeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is the procedure of choice for investigation of pyogenic spine infection (PSI) when other imaging fails to provide a de?nitive diagnosis, and for assessment of treatment response when patients are not responding as expected to antibiotic therapy [1]. Interpretation criteria for 18F-FDG PET/CT in the initial diagnosis of spine infection have previously relied on the intensity of the 18F-FDG activity. Sites with activity greater than in normal marrow are generally considered actively infected. Results from different series are dif?cult to compare because of different imaging techniques and inclusion criteria. PET-only systems (without the bene?t of a localization CT) were used prior to the availability of hybrid PET/CT. Patient populations were heterogeneous and included spine and peripheral osteomyelitis cases, infection with pyogenic organisms and tuberculosis, and non-operated and postoperative patients with and without hardware. Despite these shortcomings, a review of the early literature [2] and more recent series [3e6] have all reported high sensitivity and speci?city in the diagnosis of spine infection. The localization of 18F-FDG is nonspeci?c and occurs with infection but also with non-septic in?ammation, fracture, or neoplasm. The references quoted above have consisted of random cases, or prospective series of cases, of suspected spine infection. The prevalence of 18F-FDG-avid non-septic conditions in these series is likely low, explaining the relatively high speci?cities reported. False-positive 18F-FDG PET/CT using intensity-based criteria may be more common in problem cases done for initial diagnosis because imaging other than 18F-FDG PET/CT failed to provide a de?nitive diagnosis. Non-septic activity is commonly seen if 18F-FDG PET/CT is done for assessment of antibiotic treatment response for PSI.