Purpose: We evaluated (18)F-fluoride PET/CT for the diagnosis of screw loosening after intervertebral fusion stabilization and compared the results with those from functional radiography.
Methods: A group of 59 patients with pain in the region of previous intervertebral fusion stabilization and suspicion of implant instability due to screw loosening were investigated with (18)F-fluoride PET/CT and functional radiography, 30.1 ± 3.4 and 29.3 ± 3.2 months, respectively, after surgery. The criterion for loosening was increased focal uptake surrounding the screw entry point and shaft. SUVmax and SUVmean were measured in a region of interest (ROI) drawn around each screw (334 screws analysed). The final diagnosis was established by surgical exploration in 27 patients and clinical follow-up after intervertebral fusion stabilization in 32 patients.
Results: Of the 59 patients, 20 were proven positive for implant failure due to screw loosening and 39 were confirmed negative. The sensitivity, specificity and accuracy of (18)F-fluoride PET/CT were 75%, 97.4% and 89.8% in the patient-based analysis, and 45.6%, 100% and 80% in the screw-based analysis, respectively. The positive and negative predictive values were 93.8% and 100 % in the patient-based analysis, and 88.4 and 76% in the screw-based analysis, respectively. CT signs in PET/CT allowed screw breakage to be detected in three patients. SUVmax, SUVmean and SUVmax/SUVmean ratios in screw ROIs and respective values in reference regions were all found to be significantly different between screws positive for loosening (58 screws) and screws negative for loosening (276 screws). The ratio between SUVmax in screw ROIs and the values in reference regions was the most significant parameter for distinguishing screws positive and screws negative for loosening.
Conclusion: (18)F-Fluoride PET/CT imaging is useful for the diagnosis of screw loosening in patients with persistent symptoms after intervertebral fusion stabilization.