Introduction: Prospective bone mineral density studies after THA were conducted using dual X-ray absorptiometry (DEXA). Nevertheless, limitations of the DEXA method in contrast to computerized tomography (CT) scans have been laid bare. CT provides high resolution 3D measurements with circumferential detection of bone structures. The objective of the study presented here is the collection of prospective 5 years volumetric CT density data after cemented femoral stem implantation.

Method: The current project is based on a computerized tomography (CT) in vivo data-set of six patients. It is a five years prospective follow-up compared to the situation two years after THA (Marburg system, Centerpulse) and to the postoperative one. The 3D-analyses were done with a osteodensitometric procedure, which examines the density of each voxel in Hounsfield units (HU).

Results: The results (five years compared to postoperative) of all regions (Gruen zones) except of ROI 1 and ROI 7 demonstrated a statistically significant decreased density with p values: ROI 1 (p=0.62), ROI 2 (p=0.014), ROI 3 (p=0.023), ROI 4 (p=0.023), ROI 5 (p=0.014), ROI 6 (p=0.014), ROI 7 (p=0.3). The density reduction was greatest within ROI 2 and 3 at the lateral side of the femur.

Discussion: Bone loss of the cemented stem tested here appears to be slightly stronger than bone loss after implantation of an anatomically adapted cemented stem. To our knowledge, this is the first collection of fully prospective 5 years 3D periprosthetic density data. The volumetric CT data are superior to 2D DEXA densitometry and can be directly transferred to finite element meshes. They can be graphically post-processed in order to obtain cross-sectional or 3D displays of density patterns.