Introduction: A collar can be defined as any projection from the surface of the proximal third of the femoral stem that interferes with the capacity of the stem to move distally within the cement mantle and provide optimal load distribution along the calcar area. Contraversy exists concerning the usage of a collared or collarless prosthesis and the ability of the collar to perform its effect on the medial femoral neck. The purpose of this study is to compare the proximal femoral bone resorption and aseptic loosening in cases that had poor or good contact between the collar and the proximal medial femoral neck.

Materials& Methods: 102 hybrid total hip arthroplasties which were done for severe hip joint disease were analyzed radiographically in the current study. Pre-operative, immediate post operative and the last follow-up anteroposterior and lateral pelvis radiographies were examined. The medial femoral neck-collar contact was considered to be ideal if the medial femoral neck was fully covered by the prosthesis (group A). Contact was deamed to be poor if the medial femoral neck was partially uncovered (group B) or there was cement interposition between the bone and the prosthesis (group C) Cortical femoral bone thickness of the femoral neck was measured on the immediate post-operative and the latest follow-up radiograph as well as the thicker area of bone in Gruen Zone 7. The medial femoral neck height was measured from the superior border of the lesser trochanter

Results: The mean follow up was 4.86 years. The good contact between the collar and the medial femoral neck was achieved for most of the patients. 55 hips were in group A. 30 hips were in group B and 17 hips were in group C at the latest follow-up. The mean loss in the height of the medial femoral neck was 4.21 mm for group A, 4.26mm for group B and 3.05mm for group C. The difference among the groups was not statiscally significant (p=0.545). As we evaluate the relation between the transverse bone loss in the Gruen Zone 7; the loss was 2.49 for group A, 2.26 for group B and 1.58 for group C. The difference among the groups was not statistically significant (p=0.246)

Discussion: Unloading of the proximal femur leads resorption of the medial femoral neck and the proximal support of the prosthesis can be lost ultimately leading to aseptic failure due to excessive stresses on the proximal cement and debonding. Although the main purpose of using a collared femoral stem is to transfer load to the medial femoral neck and to prevent bone resorption, in the current study the collar did not prevent calcar resorption even when ideal contact was achieved between the collar and proximal medial femoral neck. Revision rate seems to be unchanged whether the collar had good or bad contact.