[18F]FDG PET/CT in non-union: improving the diagnostic performances by using both PET and CT criteria

Purpose Complete fracture healing is crucial for positive patient outcome. A major complication in fracture treatment is non-union. Infection is among the main causes of non-union and hence of osteosynthesis failure. For the treatment of non-union, it is crucial to understand whether a fracture is not healing because of an underlying septic process, since the surgical approach to non-unions definitely differs according to whether the fracture is infected or aseptic. We aimed to assess the diagnostic performance of 2-deoxy-2-[ 18 F]fluoro-D-glucose positron emission tomography-computed tomography ([ 18 F]FDG PET/CT) in the evaluation of infection as possible cause of non-union. Methods We retrospectively evaluated images of 47 patients treated in our trauma center who, between January 2011 and June 2017, underwent preoperative [ 18 F]FDG PET/CT aiming to exclude infection in non-union. Clinical data, diagnostic examinations, laboratory and microbiology results, and patient outcome were collected and analyzed. [ 18 F]FDG PET/CT images were visually and semiquantitatively evaluated using the maximum standardized uptake value (SUV max ). Imaging findings, as assessed by an experienced nuclear medicine physician and an experienced musculoskeletal radiologist, were compared with intraoperative microbiological culture results, which were used for final diagnosis (reference standard). The diagnostic performance of [ 18 F]FDG PET/CT in detecting infected non-union was assessed. Results Twenty-two patients were not infected, while the remaining 25 had positive intraoperative microbiological results. C-reactive protein (CRP) was within the normal range in 13 cases (five with a final diagnosis of infection) and higher than normal in 25 patients (13 with a final diagnosis of infection). Infection was correctly detected on visual analysis of PET/CT images in 23 cases, while 2/25 infected patients had no significant [ 18 F]FDG uptake and were considered false negatives. In seven cases, [ 18 F]FDG PET/CT showed false positive results; 15/22 disease-free patients were correctly diagnosed. The diagnostic accuracy of [ 18 F]FDG PET/CT in the final diagnosis of infection was 81% (38/47); its sensitivity, specificity, positive predictive value, and negative predictive value were 92%, 68%, 77%, and 88% respectively. The likelihood ratio for a positive test (LR+) was 2.89 and for a negative test, 0.12. Pretest probability of disease was 53%. Post-test probability based on LR+ wa