Background There are no reports discussing anatomic distribution of basivertebral foramen (BVF) in the osteoporotic vertebral body, which is critical in the analysis of the risk of epidural cement leakage (ECL) after cement-augmented pedicle screw fixation (CAPSF). Methods 371 osteoporotic patients using 1898 cement-augmented screws were included. Preoperative computed tomography (CT) was used to determine the frequency, width, height, and depth of magistral BVF in T10 similar to L5. Additionally, we measured the distance between BVF and the left/right borders of vertebral body as well as the distance between BVF and upper/lower endplates. Following CAPSF, the severity of ECL and the position of pedicle screws were determined by postoperative CT. Finally, significant risk factors for extensive ECL were identified through binary logistic regression analysis. Results Of 2968 vertebral bodies ranging from T10 to L5, 801 (42.2%) had a magistral BVF. From T10 to L5, the frequency of magistral BVF appeared to gradually increase. The magistral BVF was much closer to the upper endplate and the depth accounted for about a quarter of anteroposterior diameter of vertebral body. Overall, there were 19 patients (5.1%) and 32 screws (1.7%) with extensive ECL, nine of whom had neurological symptoms. The independent risk factors for extensive ECL were the magistral BVF (OR = 8.62, P < 0.001), more volume of cement injected (OR = 1.57, P = 0.031), reduced distance from screw tip to vertebral midline (OR = 0.76, P = 0.003) and vertebral posterior wall (OR = 0.77, P < 0.001) respectively. Conclusion When planning a CAPSF procedure, it is important to consider anatomical distribution of BVF and improve screw implantation methods.