Objective:To investigate the combination of internal fixation for periprosthetic fractures of the proximal femur (PFFF) after hip arthroplasty.Methods:The data of 58 patients with periprosthetic fractures after hip arthroplasty from May 2008 to March 2022 were retrospectively analyzed, including 31 males and 27 females. The average age was 75.5±18.2 years (range, 35-95 years). There were 39 total hip arthroplasty and 19 hemiarthroplasty; 37 biological prosthesis and 21 cemented prosthesis. Intraoperative periprosthetic fractures occurred in 6 cases and 52 cases postoperatively. Unified classification system (UCS): UCS IV.3A1 type 2 cases, 3A2 type 1 case, 3B1.1 type 19 cases, 3B2.1 type 25 cases, 3B3 type 2 cases, 3C type 9 cases. Fracture site: 3 cases in zone A (greater trochanter), 46 cases in zone B (around the femoral stem), and 9 cases in zone C (distal to the tip of the femoral stem. Internal fixation is composed of primary and secondary fixation, the main fixation method was the cerclage of steel wire or titanium cable, locking compression plate, and locking attachment plate fixation. The secondary fixation method was the cerclage of titanium cable, which was required to cover three zones A, B and C to form an overall balanced fixation. The modified Harris hip scores (mHHS), plate length, working length and screw number of different internal fixation combinations were compared.Results:The follow-up time was 54.2±21.6 months (range, 11-86 months). All patients showed signs of fracture healing at 10.2±1.5 weeks (range, 7-13 weeks) after operation, and bony union was observed at 19.6±1.3 weeks (range, 17-22 weeks) after operation. No delayed union or nonunion was observed. After operation, one case had a stress fracture and was revised with double-plate internal fixation; one case had a failed internal fixation and was revised with double-plate internal fixation and a large allograft bone graft. The mHHS score of UCSIV.3B2.1 group (80.3±4.6) was the lowest at 6 months after operation, and the difference between the groups of different types was statistically significant ( F=256.72, P<0.001). The score of simple internal fixation group (91.6±4.2) was higher than that of revision combined with internal fixation group (81.9±4.1), and the difference was statistically significant ( t=8.32, P<0.001). The plate length and working length were 24.9±2.5 cm and 12.6±1.7 cm for UCS IV.3B1.1, 25.4±2.6 cm and 13.6±1.8 cm for 3B2.1 and 28.1±2.5 cm and 4.9±1.9 cm for 3C, respectively ( F=5.33, P=0.005; F=6.78, P<0.001). The number of screws in zone A was significant difference among different UCS types ( F=52.67, P<0.001); UCS IV.3B1.1 (6.5±2.3) and 3B2.1 (6.7±2.2) were more than 3B3 (3.5±1.5) and 3C (3.7±1.6). The number of screws in zone B was significant difference among different UCS types ( F=42.15, P<0.001); The number of UCS IV.3B1.1 (2.3±1.6) and 3B2.1 (2.8±1.9) were significantly more than that of 3B3 (1.0±0.5) and 3C (1.2±0.6). The number of screws in zone C was significant differences among different UCS types ( F=39.62, P<0.001); The number of UCS IV.3B1.1 (3.8±1.9) and 3B2.1 (3.9±1.7) were more than that of 3B3 (2.0±0.5), the difference was statistically significant ( P<0.05). Conclusion:The function of hip after simple internal fixation of proximal femoral periprosthetic fractures was better than that of those who underwent revision at the same time; the number of screws of UCSIV.B1 and B2 is more than that of B3.