ObjectiveCT three-dimensional reconstruction technology was used to simulate the placement of the lumbar cortical bone trajectory (CBT), to determine the starting point and direction of the screw trajectory.MethodsBetween February 2017 and April 2018, 24 patients with lumbar CT were selected as the study object. There were 7 males and 17 females, with an average age of 50.4 years (range, 37-68 years). The CT DICOM data of patients were imported into Mimics 16.0 software, and the three-dimensional model of lumbar spine was established. A 5 mm diameter cylinder was set up to simulate the CBT by using Mimics 16.0 software. According to the different implant schemes, the study was divided into groups A, B, and C, the track of the screw respectively passed through the upper edge, the medial edge, and the lower edge of the isthmus of the pedicle. The intersection of simulated screw and lumbar spine was marked as region of interest (ROI) and a mask was generated. The average CT value [Hounsfield unit (HU)] and the screw length of ROI were automatically measured by Mimics 16.0 software. In addition, the head inclination angle and head camber angle of the screw were measured respectively. Point F was the intersection of the level of the lowest edge of the transverse process and the lumbar isthmus periphery. The horizontal and vertical distance between point F and the starting point were measured, and the relationship between the three schemes and the position of the zygapophysial joint and spinous process was observed.ResultsPlan A has the highest ROI average HU, with the maximum value appearing in L4; plan B has the longest screw length, with the maximum value appearing in L5; plan C has the largest nail track head inclination angle, with the maximum value appearing in L4; plan B has the largest nail track head camber angle, with the maximum value appearing in L3. The screw length and head camber angle of the nail in group B were significantly greater than those in groups A and C (P<0.05); the head inclination angle in groups A, B, and C was gradually increased, showing significant differences (P<0.05); there was no significant difference in the average HU value of ROI between the 3 groups (P>0.05). In plan A, 74.48% (143/192) screws had a horizontal distance of −2 to 4 mm from point F, a vertical distance of 6-14 mm from point F, a head inclination angle of (14.64±2.77)°, and a head camber angle of (6.55±2.09)°, respectively; in plan B, 84.58% (203/240) screws had a horizontal distance of 1-6 mm from point F, a vertical distance of 1-5 mm from point F, a head inclination angle of (26.93±2.21)°, and a head camber angle of (10.29±2.46)°, respectively; in plan C, 85.94% (165/192) screws had a horizontal distance of −2 to 3 mm from point F, a vertical distance of −2 to 4 mm from point F, a head inclination angle of (33.50±3.69)°, and a head camber angle of (6.47±2.48)°, respectively.ConclusionPlan B should be selected as the starting point of the L1-L5 CBT implant. It is located at the intersection of the lowest horizontal line of the transverse process root and the lateral edge of the lumbar isthmus, which is 1-6 mm horizontally inward, 1-5 mm vertically upward, with a head inclination angle of (26.93±2.21)°, and a head camber angle of (10.29±2.46)°, respectively.