Blood supply of the proximal metaphysis of the femur comes mainly from the lateral circumflex femoral artery (LCFA). Essentially, the anterior approach has gained popularity in hip surgery but routinely requires the ligation of the ascending branch of the LCFA. Until now, there is no study analysing the effect of previous hip surgery on the vascularization of the proximal femur. Notably, it might, however, have consequences on osteointegration of uncemented prosthesis as well as in the management of early complications. Therefore we conducted a retrospective study to address the following questions: (1) Is the blood supply of the trochanteric region impaired by previous hip surgery, (2) does the anterior approach alter it more than other ones?

We hypothesised that the surgical approach to the hip influences blood supply of the proximal femur, as visualised by retrospective analysis of femoral digital subtraction arteriograms (FDSA).

A retrospective review of 1280 FDSA, performed for vascular indications with a standard frame rate, between 07/2014 and 06/2016 in a single institution. Qualitatively sufficient FDSA were divided into 4 groups according to the history of previous hip surgery: hip replacement through an anterior approach (n=10) or through a lateral approach (n=31), cephalomedullary nailing for fractures of the proximal femur (n=5), and a control group of 50 continuous patients without previous hip surgery. The number of frames was counted between contrast injection into the femoral bifurcation and filling of the ipsilateral vessels of the greater trochanter to measure a potential delay/impairment of its arterial perfusion. Anatomic variations of the LCFA were also recorded.

The number of frames between contrast injection and visualisation of the blood supply of the greater trochanter was 3.6±0.9 (mean±SD) in the control group (p<0.001 vs. all other groups). In patients with a hip replacement, the delay was 7.0±1.9 frames for the anterior approach and 5.2±1.1 frames for the lateral approach, respectively. In patients after cephalomedullary nailing, a delay of 4.8±1.5 frames was measured. The delay in the anterior approach group was significantly longer (p<0.001) compared to all other investigated groups. The ascending branch of the LCFA could not be detected after the anterior approach. As after lateral approach or as in the control group, the transverse branch was detectable in approximately 2/3 of the patients.

Arterial perfusion of the greater trochanter is impaired after hip surgery, particularly after an anterior approach. The clinical relevance of these findings still needs to be investigated. It might, however, explain some early aseptic failures of uncemented stems. Moreover, it should be considered in early revision surgery, because combining different approaches might critically impair femoral blood supply.

III, retrospective case control study.