Abstract: Objective： To establish a reliable approach for measuring proximal femoral 3 dimensional anatomy, and to compare post-operative differences of proximal femoralanatomy in the inter-trochanter fractures with two kinds of antegrade nailings.Methods: Some computer assisted design (CAD) softwares, e.g. Mimics, were used to establish a reliable approach for measuring proximal femoral 3 dimensional (3D) anatomy. Intra-class correlation coefficient (ICC) was used to test the reliability of intra- and inter-observers. The post-operative pelvic CT data of 19 cases of inter-trochanter fracture patients treated with InterTAN nailing and 21 cases of inter-trochanter fracture patients treated with proximal femoral nail anti-rotation (PFNA) were retrospectively analysed and used to measure bilateral proximal femoral anatomical parameters, including 2D and 3D femoral neck-shaft (NS) angle and femoral neck anteversion (NA) angle, and 2D and 3D anteversion angles of the intramedullary (IM) nailings. ICC was used to test the consistency of the NA angles in the different groups, and the paired student T-test was used to test the differences of the paired quantitative data. Results: The established measurement method hasdexcellent consistency within the intra- and inter-observers, with all the ICCs higher than 0.9. The paired student T-test showed no significant difference between the post-operative bilateral 2D or 3D NA angles. The ICCs results showed that there were no consistency between the post-operative bilateral 2D or 3D NA angles (P values were 0.099 and 0.055, respectively), but the excellent consistency between the 2D injured side NA angle and 2D IM nailing’s NA angle, or between the 3D injured side NA angle and 3D IM nailing’s NA angle (the ICCs were 0.81 and 0.8, respectively, P values < 0.001). In PFNA group, 57% of the differences between the 2D post-operative injured side’s and intact side’s NA angles were higher than 15°, which was more than 15.78% in InterTAN group. The paired student T-test showed no significant difference between the post-operative injured side’s 2D or 3D NS angles and the intact side’s respective 2D or 3D angles in PFNA group (P values were 0.925 and 0.367, respectively), but in InterTAN group, the post-operative injured side’s 2D or 3D NS angles were significantly smaller than the intact side’s respective angles (P values were 0.033 and 0.009, respectively). Conclusion: By analyzing and comparing bilateral proximal femoral anatomical parameters after two kinds of IM nailings procedures, the differences between the bilateral post-operative NA angles in PFNA group were significantly larger than those in InterTAN group. There was significant correlation between the NA angles of the injured sides and NA angles of IM nailings in both the groups.

Key words: Femoral fractures, Fracture fixation, intramedullary, Bone nails, Computer-aided design