Craniocaudal cyclic load improve risk assessment of lumbar pedicle screw loosening：finite element analysis based on computer tomography

Abstract Background: Screw loosening remains a prominent complication for osteoporotic patients with pedicle screw fixation surgeries, yet with limited risk assessment approach. The aim of this study was to investigate influence of craniocaudal cyclic load on pedicle screw fixation strength by computed tomography (CT) based finite element analysis (FEA) and we examined predict ability in pedicle screw loosening (PSL). Methods: 12 clinical PSL cases (7 men, 5 women) and 12 age- and sex-matched controls were enrolled for CT based FEA. Simple axial pullout load and axial pullout load with preset craniocaudal cyclic load were applied to each model respectively, and the ultimate pullout force under both conditions is calculated as the fixed strength and compared. Besides, HU values of the vertebral body trabeculae and screw trajectory were measured as an assessment of osteoporosis. The ultimate pullout force and HU value were compared between PSL and controls cases. Results The cyclic load remarkably reduce the pullout force of pedicle screws (906.2 ± 180.2 N vs. 729.3 ± 172.3 N, p<0.0001) by CT based FEA. No significant difference between the PSL and the control group in the simple axial pull-out force and HU values of the vertebral body. But the pullout force with preset cyclic load (639.2 ± 169.4 N vs. 819.4 ± 125.1 N, p = 0.072) and the HU value of the screw trajectory (177.5 ± 43.8 vs. 217.2 ± 29.6, p = 0.016) in the PSL group is significantly lower than that in the control group. Area under receiver operating characteristic curve (ROC) revealed pullout force with preset cyclic load slightly better predicted PSL than HU value of the screw trajectory (AUC = 0.798 vs. 0.750). Conclusions The craniocaudal cyclic load significantly reduces the screw fixation strength. HU value of screw trajectory and pullout force with preset cyclic load by CT based FEA are helpful for the clinical prediction of PSL.