J Bone Miner Metab (2003) 21:229–233 Sequential changes in periprosthetic bone mineral density following total hip arthroplasty: a 3-year follow-up Hiroshi Ohta, Seneki Kobayashi, Naoto Saito, Masashi Nawata, Hiroshi Horiuchi, and Kunio Takaoka Department of Orthopaedic Surgery, Shinshu University, School of Medicine, 3-1-1 Asahi, Matsumoto 390-8621, Japan sthetic bone mineral density (BMD), and a number of reports on changes in BMD as measured by DXA have been published [5–8]. However, only two longterm, longitudinal periprosthetic BMD studies exist [9,10]. Furthermore, there is no consensus as to which factors affect periprosthetic BMD. The purpose of this study was to measure changes in periprosthetic BMD over a 3-year period and to identify which factors were respon- sible for any change. Patients and methods Twenty-one patients participated in the study. There were three men and 18 women, with an average age of 59.8 years (range, 38–79 years). Patients were operated on at Shinshu University Hospital from November 1995 to August 1998, using standard operative techniques. The preoperative diagnosis was osteoarthritis in 18 hips, and osteonecrosis of the femoral head in 3 hips. The hip prosthesis used in this study was the Spongiosa Metal Total Hip system (S  G Implants; Lübeck, Germany), a cobalt-chrome-molybdenum alloy with a full porous coating that has a spongy metal structure, similar to cancellous bone, on its surface (Fig. 1). Patients were allowed partial weight-bearing beginning 2 weeks after the operation, and full weight-bearing after 2 months. No patients received anti–bone–resorption therapy, such as etidronate or alendronate, and no women re- ceived hormone replacement therapy. All three cases of osteonecrosis of the femoral head were complications of steroid use. Corticosteroid therapy was discontinued preoperatively. Changes on radiographs were also examined during the follow-up period. Radiologic fixation failure of the femoral prosthesis was defined as aseptic loosening (progressive subsidence and at least partially sur- rounded by divergent radio-opaque lines that were more widely separated from the stem at its extremities) Abstract We sequentially measured the periprosthetic bone mineral density (BMD) of the femur after cementless total hip arthroplasty, using dual-energy X-ray absorptiometry, over a 3-year period. The periprosthetic bone was divided into three regions (proximo-medial, middle, and distal to the prosthetic stem). After the insertion of a fully porous coated stem in 21 patients, the BMD was measured within 3 weeks, and 6, 12, 24, and 36 months after surgery. At 6 months, all zones showed a decrease in BMD relative to the BMD within 3 weeks, but subsequently the BMD was unchanged. The lower the BMD within 3 weeks of surgery, or the lower the body weight, the higher the percent loss of BMD at 6 months. Key words total hip arthroplasty · bone mineral density · dual-energy X-ray absorptiometry · initial postoperative bone mineral density · body weight Introduction In the normal proximal femur, load is transmitted from the femoral head through the cortical and cancellous bone of the femoral neck, to the femoral diaphysis. When a metallic implant is inserted, loading conditions on the femur change [1]. That is, the body weight is loaded directly onto the diaphysis, by passing the proxi- mal femur. This change in load can lead to resorption of femoral bone. Stress shielding is believed to be the main cause of this bone loss [2,3]. The problem is that such bone loss makes revision surgery difficult, and adequate bone stock must be preserved for surgical revision [4]. Therefore, periprosthetic bone loss should be moni- tored after total hip arthroplasty (THA). Dual-energy X-ray absorptiometry (DXA) can ex- press the sequential postoperative changes in peripro- Offprint requests to: H. Ohta Received: September 9, 2002 / Accepted: December 12, 2002 © Springer-Verlag 2003