A study of screw placement to obtain the optimal pull-out resistance of lumbar pedicle screws—analysis of Hounsfield units measurements based on computed tomography

Abstract

Abstract Objective The screw path of lumbar pedicle screws in the vertebral body has certain variability. It is not clear whether the screw paths in different directions can obtain the same pull-out resistance. This study intends to use CT (Computed Tomography) to measure the Hounsfield unit (HU value) around the screw paths in different parts of the lumbar vertebral body to obtain the bone mineral density value of the corresponding parts which will provide some reference for the direction of lumbar pedicle screw placement. Methods This retrospective study included 200 patients with lumbar degenerative diseases selected randomly from the case base and the patient’s basic information was recorded. L1-L5 vertebral body was divided equally into the upper, middle and lower 1/3, which was consistent with the three sagittal entry directions of the pedicle screw head tilt, parallel endplate and caudal tilt, and the HU values were measured by CT cross-sectional scanning to indirectly reflect the local bone density values. The paired t-test (randomized block experiment) was used to compare the HU values of the upper, middle and lower 1 / 3 parts, with P < 0.05 being considered statistically significant. Results Comparison of HU values in different parts of each vertebral body revealed that HU values in the middle 1/3 of the L1,L2 (163.88 ± 58.44 and 152.94 ± 59.45) and in the lower 1/3 of the L4 (149.86 ± 60.18) were higher than in the other two parts of the vertebral body of the same segment(P < 0.0001,P = 0.0069 and P = 0.0024, respectively); According to the results of each stratification, patients with younger age and better bone condition had higher HU values in the middle 1/3 of L1 and L2, and higher HU values in the lower 1/3 of L3, L4 and L5; With the increase of age, the decrease of bone condition and the difference of HU value in each vertebral body gradually decreased. Conclusion Although further follow-up studies are needed, based on the analysis of the statistical results, we speculate that from the perspective of obtaining the best pull-out resistance of the lumbar pedicle screws, the placement direction of L1 and L2 in the sagittal position may be as parallel to the endplate as possible; L3, L4, and L5 may be as appropriate as possible to the tail tilt theoretically.