Can Spatiotemporal Fluoride (18F−) Uptake be Used to Assess Bone Formation in the Tibia? A Longitudinal Study Using PET/CT

Background When a bone is broken for any reason, it is important for the orthopaedic surgeon to know how bone healing is progressing. There has been resurgence in the use of the fluoride ( 18 F − ) ion to evaluate various bone conditions. This has been made possible by availability of positron emission tomography (PET)/CT hybrid scanners together with cyclotrons. Absorbed on the bone surface from blood flow, 18 F − attaches to the osteoblasts in cancellous bone and acts as a pharmacokinetic agent, which reflects the local physiologic activity of bone. This is important because it shows bone formation indicating that the bone is healing or no bone formation indicating no healing. As 18 F − is extracted from blood in proportion to blood flow and bone formation, it thus enables determination of bone healing progress. Questions/purposes The primary objective of this study was to determine whether videos showing the spatiotemporal uptake of 18 F − via PET bone scans could show problematic bone healing in patients with complex tibia conditions. A secondary objective was to determine if semiquantification of radionuclide uptake was consistent with bone healing. Methods This study investigated measurements of tibia bone formation in patients with complex fractures, osteomyelitis, and osteotomies treated with a Taylor Spatial Frame TM (TSF) by comparing clinical healing progress with spatiotemporal fluoride ( 18 F − ) uptake and the semiquantitative standardized uptake value (SUV). This procedure included static and dynamic image acquisition. For intrapatient volumes acquired at different times, the CT and PET data were spatially registered to bring the ends of the bones that were supposed to heal into alignment. To qualitatively observe how and where bone formation was occurring, time-sequenced volumes were reconstructed and viewed as a video. To semiquantify the uptake, the mean and maximum SUVs (SUVmean, SUVmax) were calculated for the ends of the bones that were supposed to heal and for normal bone, using a spherical volume of interest drawn on the registered volumes. To make the semiquantitative data comparable for all patients with multiple examinations, the SUVmean and SUVmax difference per day (SUVmeanDPD and SUVmaxDPD) between the first PET/CT scan and each subsequent one was calculated. Indicators of poor healing progress were (1) uneven distribution of the radionuclide uptake between ends of the bones that were supposed to heal as seen in the video or, (