Changes in proximal femoral bone mineral density around a hydroxyapatite-coated hip joint arthroplasty

Journal of Orthopaedic Surgery, Jun 2003 by Theis, J C, Beadel, G

ABSTRACT

Objective. To present the results of a prospective analysis of proximal femoral bone mineral density changes around a hydroxyapatite-coated total hip joint replacement.

Methods. 14 patients with osteoarthritis of the hip were enrolled in the study and treated with an uncemented ABG prosthesis. Dual energy X-ray absorptiometry scanning was performed in 9 patients preoperatively, and at 3, 6, 12, and 24 months postoperatively. An orthopaedic software program was used to determine the bone mineral density in the proximal femur, expressed as a percentage of the preoperative value.

Results. The values of Gruen zones 1 to 6 averaged between 96.0% and 113.8% of the preoperative value by 24 months (overall average, 104.1%). In zone 7, however, there was a gradual decline in bone mineral density to an average of 72.1% of the preoperative value by 24 months. This represented ongoing loss of bone from the calcar; although this may not pose a problem to the prosthetic's short-term stability, it may render potential revision surgery more difficult.

Conclusion. The initial outcome of uncemented total hip replacement appears to be promising. There was excellent maintenance of bone around the femoral component in all regions other than the calcar and lesser trochanter. Further scans are required to see if these trends continue in the long term.

Key words: arthroplasty, replacement, hip; densitometry; hydroxyapatite

INTRODUCTION

Changes of bone mineral density (BMD) around total hip joint replacements, particularly around the femoral component, are of great interest because periprosthetic bone loss increases the risk of fixation loss, implant migration, and periprosthetic fracture.

Many published studies document BMD changes around various implants, showing significant bone loss that increases with time,1,2 appears worse in women than in men, and occurs to a greater extent around the proximal rather than distal part of the femoral stems.3 Potential causes include age-related bone loss, stress shielding, and possibly other factors.4,5

The extent of periprosthetic bone loss at time of detection may predict long-term clinical success or failure. With early detection, therapy such as the use of biphosphonates and other drugs can prevent significant bone loss. Radiography is insensitive to early-stage bone loss, and any loss of less than 70% is not reproducibly detected.2 Quantification is also difficult, but dual energy X-ray absorptiometry (DEXA) has been widely used to measure BMD around joint implants. DEXA scanning offers the advantages of short scanning time (2-5 minutes), and a low radiation dose (0.5-5.0 mSv), which is less than 5% the dose of a chest X-ray. Such characteristics make DEXA ideal for repeated examinations in the same patient.6,7

Several studies8-10 have shown the accuracy of DEXA scanning to be excellent. The use of phantoms has shown accuracy and precision errors to occur in less than 1% in vitro and less than 5% in vivo.8 Another study using hydroxyapatite phantoms has shown a coefficient of variation of less than 2%, and a mean position error of 2.7%-3.4% in vivo, with the most significant contributing factor being changes in rotation.9 However variation of rotation between 15[degrees] internal and 15[degrees] external rotation has been reported to cause an error of measurement of less than 5%.10

The aims of our study were to analyse femoral BMD changes around a hydroxyapatite-coated total hip arthroplasty, and to measure periprosthetic change of bone mass in a 2-year period.

MATERIALS AND METHODS

14 patients were enrolled in this study. All the patients had hip osteoarthritis. They had given informed consent before the study. Each patient received an uncemented, hydroxyapatite-coated total hip replacement (ABG; Stryker Howmedica, New Jersey, United States). The system has a titanium anatomic stem coated with hydroxyapatite in the proximal metaphyseal region only. Arthroplasty was performed by a number of surgeons using either a lateral or posterior approach. Postoperatively, patients were mobilised, partially weightbearing for the first 6 weeks, and thereafter fully weightbearing. Each subject was managed according to a local protocol for total joint arthroplasty.

DEXA scanning was performed on each patient preoperatively, and again at 3, 6, 12, and 24 months postoperatively. A Lunar DPX-L scanner (GE Medical Systems, Fairfield, United States) was used to obtain scans of each patient in a supine position, with the leg held in a neutral position by a foam knee-positioning device, toes to the ceiling, and the foot strapped into a rigid holder in order to reduce rotational variation between scans. Rice bags were placed on the outside of each patient's thigh to ensure that the scanner did not pass through air.

The periprosthetic femoral bone mineral densities were determined for each of the 7 Gruen zones (Fig. 1) and for 4 regions of interest (Fig. 2) within the proximal femur, using an orthopaedic software package (GE Medical Systems, Fairfield, United States) with metal-removal function supplied by the manufacturer.