Objective To investigate the effectiveness of three-dimensional (3D)-printed hemi-pelvic prosthesis for revision of aseptic loosening or screw fracture of modular hemi-pelvic prosthesis. MethodsBetween February 2017 and January 2020, 11 patients with aseptic loosening or screw fracture of modular hemi-pelvic prosthesis were revised using 3D-printed hemi-pelvic prostheses. There were 7 males and 4 females with an average age of 44 years (range, 25-60 years). In the first operation, all patients underwent total tumor resection, modular hemi-pelvic prosthesis reconstruction, and autologous femoral head transplantation. According to the Enneking pelvic partition system, 8 cases were resected in zones Ⅰ+Ⅱ and 3 cases in zones Ⅰ+Ⅱ+Ⅲ. The interval from the initial operation to this revision ranged from 14.3-66.2 months, with an average of 35.8 months. The operation time, the amount of intraoperative bleeding, and the occurrence of complications were recorded. At 6 months after the first operation, before revision, and at last follow-up, the American Musculoskeletal Tumor Society (MSTS) score and Harris score were used to evaluate the recovery of lower limb function. The pain-free walking distance of patients without brace assistance was recorded at last follow-up. X-ray films were taken at 1 month after the first operation, before revision, and at 1 month after revision, the acetabulum position was assessed by the differences in weight arm and cup height between bilateral hip joints. At last follow-up, the digital X-ray tomography was taken to evaluate the prosthesis-bone integration and the occurrence of aseptic loosening. Results The operation time was 182.6-238.0 minutes (mean, 197.4 minutes). The amount of intraoperative bleeding was 400-860 mL (mean, 550.0 mL). All incisions healed by first intention with no infection, hip dislocation, nerve damage, or vascular-related adverse events. The MSTS score and Harris score at last follow-up were significantly higher than those at 6 months after the first operation and before revision (P<0.05), while the score before revision was significantly lower than that at 6 months after the first operation (P<0.05). At last follow-up, the patients were able to walk more than 1 000 meters painlessly without brace assistance. Imaging review showed that the difference of cup height at 1 month after revision was significantly lower than that at 1 month after the first operation and before revision, and at 1 month after the first operation than before revision operation, and the differences were significant (P<0.05). There was no significant difference in the difference of weight arm among three time points (P>0.05). All prostheses were well integrated, and no aseptic loosening of the prosthesis or screw fracture occurred. Conclusion Revision with 3D-printed hemi-pelvic prostheses benefited in reconstructing stable pelvic ring and natural bodyweight transmission for patients encountering the aseptic loosening or screw fracture of modular hemi-pelvic prosthesis. Early postoperative rehabilitation training can maximize the recovery of patient limb function, reduce pain during walking, and reduce the incidence of complications.