Hydroxyapatite-coated uncemented hip stems and bone remodeling

Abstract

In this thesis the clinical results, the periprosthetic bone remodeling and histological analysis of an anatomical designed proximally hydroxyapatite-coated hip prosthesis were investigated to answer several research questions. In our first prospective study the characteristics of the bone remodeling pattern around an uncemented anatomical, proximally coated stem (ABG-I) and its successor (ABG-II) were described using DEXA. ABG-II showed less bone loss proximally, especially in zone 7. However, the difference found for the proximal zone 7 was not significant In our second study, patients were retrospectively matched for preoperative bone quality and gender. Now, a statistically significant difference was found in proximal bone preservation in favor of ABG-II for zone 7. These findings support our conclusion that in comparative bone remodeling studies for prosthetic design using DEXA, patients should be matched for preoperative bone quality and gender to limit the number of patients while maintaining maximum statistical power. Bone remodeling can also be influenced by tightness of stem fit and therefore we assessed the radiological changes five years after implantation of the stem (third study). We investigated if there was a correlation between a tight fit and the bone remodeling pattern. In this study there was no correlation between fit and radiological changes, but there was a correlation between a poor fit and thigh pain for the ABG-II stem. In the fourth study bone remodeling was assessed by histomorphometric analysis of a proximally HA-coated stem (ABG-I). In this study the HA residue was measured histomorphometrically on hip stems of one single design (ABG-I) retrieved at post-mortem, and the long-term performance of the HA coating and the effects of resorption were investigated. HA resorption increased significantly with the time in vivo as measured by the residual HA. Beyond 8 years HA was almost gone. HA resorption was significantly more proximal than distal, an effect which was less strong with bone ongrowth. Bone ongrowth was statistically independent of the time in-vivo. Bone ongrowth was significantly less with older patients. In the fifth and last study periprosthetic fractures around an uncemented proximally coated stem are described. The fracture patterns were analyzed by reviewing the X-rays in detail and the modified Vancouver classification was retrospectively applied for these fractures. An increased number of periprosthetic fractures for the ABG-I stem compared to the cemented stem was shown. The site of the fracture in our series was always localized in an area where an acute bone density gradient in the cortex had developed, between the proximal and the middle Gruen zones. The retrospectively applied Vancouver classification turned out to be a simple, reproducible classification system for the uncemented treatment modality, which will make clinical decision-making easier. All studies performed point in the same direction: proximal bone loss for the ABG stem and less proximal bone loss for the ABG-II stem compared to the ABG-I stem. Despite this proximal bone resorption, clinical follow up shows excellent results after ten years for the ABG-I stem. Its successor (ABG-II) shows less proximal bone resorption, but a relatively high percentage of thigh pain. Long term follow-up studies should provide more information as to whether less proximal bone resorption (ABG-II) will also lead to better clinical performance. Stress shielding is and will remain an important issue in total hip arthroplasty for the long term.