Periprosthetic bone loss after insertion of an uncemented, customized femoral stem and an uncemented anatomical stem. A randomized DXA study with 5-year follow-up

Background and purpose

Customized femoral stems are designed to have a perfect fit and fill in the femur in order to achieve physiological load transfer and minimize stress shielding. Dual-energy X-ray absorptiometry (DXA) is regarded as an accurate method for detection of small alterations in bone mineral density (BMD) around hip prostheses. We present medium-term DXA results from a randomized study comparing a customized and an anatomical femoral stem.

Methods

100 hips were randomized to receive either the anatomical ABG-I stem or the Unique customized femoral stem, both uncemented. DXA measurements were conducted postoperatively and after 3, 6, 12, 24, 36, and 60 months, and BMD was computed for each of the 7 Gruen zones in the proximal femur.

Results

Results from 87 patients were available for analysis. 78 completed the 5-year follow-up: 35 patients in the ABG group and 43 patients in the Unique group. In both groups, we found the greatest degree of bone loss in the proximal Gruen zones. In zone 1, there was 15% reduction in BMD in the ABG-I group and 14% reduction in the Unique group. In zone 7, the reduction was 28% in the ABG-I group and 27% in the Unique group. The only statistically significant difference between the groups was found in Gruen zone 4, which is distal to the tip of the stem, with 1.6% reduction in BMD in the ABG-I group and 9.7% reduction in the Unique group (p = 0.003).

Interpretation

5-year DXA results showed that because of stress-shielding, proximal bone loss could not be avoided—either for the anatomical ABG-I stem or for the customized Unique stem.Implantation of a prosthesis in the femur alters the load distribution in the host bone, and the femur remodels to adapt to the new mechanical situation (Engh and Bobyn 1988, Bobyn et al. 1992, Engh et al. 1999). A perfect fit and fill in the proximal femur is said to be important to achieve physiological load transfer (Laine et al. 2000). It has therefore been hypothesized that a customized femoral stem could fulfill the criterion for optimal fit and fill, and thus minimize stress shielding in the proximal femur (Bargar 1989). Based on numerical analyses, however, it has been claimed that canal-filling femoral stems may cause stress shielding and subsequently bone atrophy of the proximal femur (Huiskes et al. 1989).In vitro studies on human cadaver femurs have indicated that insertion of a customized stem gives a better pattern and distribution of cortical strains in the proximal femur than anatomical stems (Aamodt et al. 2001, Østbyhaug et al. 2009). In these studies, it was shown that 33–56% of the external cortical strains were retained on the proximal, medial aspect of the femur after insertion of a customized stem whereas the corresponding figure was 10–13% after insertion of a standard, anatomical stem. Long-term bone remodeling is, however, not necessarily reflected in the immediate postoperative mechanical situation in the femur as measured in experimental strain analyses.The present randomized, clinical study was undertaken to compare (by DXA) the medium-term changes in bone mineral density in the proximal femur after insertion of an uncemented, customized femoral stem and an uncemented, standard anatomical femoral stem.