BACKGROUND: The choice of intramedullary nails versus extramedullary fixation in surgical treatment of intertrochanteric fractures remains controversial.
OBJECTIVE: To evaluate the efficacy of intramedullary versus extramedullary fixation for the treatment of intertrochanteric fractures.
METHODS: A computer-based search was performed on the MEDLINE database, EMBASE database, Cochrane Library, CBM database, CNKI database and other commonly used electronic database, supplemented by manual searches to find the randomized controlled trials on the comparison of intramedullary versus extramedullary fixation for the treatment of intertrochanteric fractures at home and abroad. All randomized controlled trials comparing the efficacy of intramedullary versus extramedullary internal fixation for the treatment of intertrochanteric fractures were included. The included studies were evaluated with the modified Jadad quality scale, and the software Stata 12.0 and RevMan 5.1 were used for data analysis. The statistical method was the Mantel-Haenszel method.
RESULTS AND CONCLUSION: A total of 31 randomized controlled trials were included, including 5 217 patients. According to the modified Jadad quality scale, 19 literatures of methodological quality showed high quality, and low quality of 12 literatures. The intraoperative femoral fractures [relative risk (RR)=4.00, 95% confidence interval (CI) (1.80-8.86), P < 0.01], postoperative femoral fracture [RR=3.57, 95%CI (1.68-7.60), P < 0.01], cut out [RR=1.61, 95%CI (1.02-2.52); P=0.04], reoperation risk [RR=1.58, 95%CI (1.09-2.30); P=0.02], and residual hip or thigh pain in the follow-up period [RR=1.31, 95%CI (1.09-1.57), P < 0.01] in the Gamma nail group was significantly higher than those in the sliding hip screw group. The intraoperative blood loss in the proximal femoral nail anti-rotation group was significantly less than that in the sliding hip screw group, and the hip joint Harris score was higher, but the intraoperative radiation time was longer [mean difference (MD)=2.11, 95%CI (1.78-2.43), P < 0.01]. The intraoperative blood loss in the intramedullary hip screw group was significantly less than that in the sliding hip screw group, but the radiation time was longer [MD=1.20, 95%CI (0.88-1.52), P < 0.01]. For stable intertrochanteric fracture, both screw-plate system and intramedullary fixation have good outcome, but the dynamic hip screw as the representative of extramedullary fixation system is preferred. The intramedullary system is better than extramedullary fixation system in the treatment of unstable intertrochanteric fractures.