The aim of this retrospective cohort study was to identify any difference in femoral offset as measured on pre-operative anteroposterior (AP) radiographs of the pelvis, AP radiographs of the hip and corresponding CT scans in a consecutive series of 100 patients with primary end-stage osteoarthritis of the hip (43 men and 57 women with a mean age of 61 years (45 to 74) and a mean body mass index of 28 kg/m² (20 to 45)).

Patients were positioned according to a standardised protocol to achieve reproducible projection and all images were calibrated. Inter- and intra-observer reliability was evaluated and agreement between methods was assessed using Bland-Altman plots.

In the entire cohort, the mean femoral offset was 39.0 mm (95% confidence interval (CI) 37.4 to 40.6) on radiographs of the pelvis, 44.0 mm (95% CI 42.4 to 45.6) on radiographs of the hip and 44.7 mm (95% CI 43.5 to 45.9) on CT scans. AP radiographs of the pelvis underestimated femoral offset by 13% when compared with CT (p < 0.001). No difference in mean femoral offset was seen between AP radiographs of the hip and CT (p = 0.191).

Our results suggest that femoral offset is significantly underestimated on AP radiographs of the pelvis but can be reliably and accurately assessed on AP radiographs of the hip in patients with primary end-stage hip osteoarthritis.

We, therefore, recommend that additional AP radiographs of the hip are obtained routinely for the pre-operative assessment of femoral offset when templating before total hip replacement.

Pre-operative computerised three-dimensional planning was carried out in 223 patients undergoing total hip replacement with a cementless acetabular component and a cementless modular-neck femoral stem. Components were chosen which best restored leg length and femoral offset. The post-operative restoration of the anatomy was assessed by CT and compared with the pre-operative plan.

The component implanted was the same as that planned in 86% of the hips for the acetabular implant, 94% for the stem, and 93% for the neck-shaft angle. The rotational centre of the hip was restored with a mean accuracy of 0.73 mm (sd 3.5) craniocaudally and 1.2 mm (sd 2) laterally. Limb length was restored with a mean accuracy of 0.3 mm (sd 3.3) and femoral offset with a mean accuracy of 0.8 mm (sd 3.1).

This method appears to offer high accuracy in hip reconstruction as the difficulties likely to be encountered when restoring the anatomy can be anticipated and solved pre-operatively by optimising the selection of implants. Modularity of the femoral neck helped to restore the femoral offset and limb length.

Intertrochanteric osteotomy may postpone the need for total hip replacement (THR). In young patients with an acquired deformity of the femoral head and secondary osteoarthritis, a valgus intertrochanteric osteotomy may allow better congruency but the acetabular cover may become insufficient because of subluxation of the femoral head. In patients with a spherical femoral head and acetabular dysplasia, cover can still remain insufficient after varus displacement osteotomy. We present the long-term results of intertrochanteric osteotomy combined with an acetabular shelfplasty in both these circumstances.

Sixteen hips (15 patients) with a deformed femoral head, and ten (seven patients) with a spherical femoral head, underwent an intertrochanteric osteotomy and acetabular shelfplasty. The mean age at the time of surgery was 30 and 37 years and the mean final follow-up was 15 and 19 years, respectively. Six patients in the deformed group, but only one in the spherical group, had required a THR by the time of their final follow-up. In both groups, those who had not undergone a THR had a good result.

Acetabular shelfplasty is an excellent addition to an intertrochanteric osteotomy and gives full cover of the femoral head in patients with a deformity of the head and secondary osteoarthritis.