2.4. Radiographic Evaluation

The cortical index was calculated from anteroposterior plain radiographs of the hip using the method described by Dorr et al. [18]. The cortical index was obtained by dividing the thickness of the femoral bone cortex at a point 10 cm distal to the center of the lesser trochanter (i.e., the length of the lateral aspect of the femoral bone cortex minus the length of the medullary cavity) by the lateral extent of the femoral bone cortex. The morphology of the proximal femur was classified as Dorr type A, B, or C [18], which corresponded to measured cortical indices of ≥0.58, 0.49–0.57, and ≤0.48, respectively, according to the description by Nakaya et al. [19]. Stem subsidence was evaluated 1, 4, and 12 weeks after surgery. Subsidence of at least 2 mm on postoperative radiographs was considered clinically significant and recorded as positive subsidence [20]. Subsidence of the femoral stem was measured as the distance between the proximal aspect of the greater trochanter and shoulder of the femoral stem [21]. The method used to measure stem subsidence is shown in Figure 3. Femoral BMD scanning was performed using a Hologic Discovery A densitometer with Apex software (Apex Version 5.6.0.4. Hologic Inc., Marlborough, MA, USA). BMD was selected in four regions of the femur: the femoral neck, trochanter, intertrochanteric region, and total femur [22].

Radiographic measurement of femoral stem subsidence. The white line (a): Subsidence was measured as the distance between the proximal aspect of the greater trochanter and shoulder of the femoral stem on calibrated anteroposterior X-rays of the hip. The vertical red line: coronal stem alignment.