Introduction: After total hip arthroplasty (THA) the periprosthetic bone is loaded in an unphysiological manner (stress shielding), a major cause for periprosthetic bone resorption and aseptic loosening. Design, material and surface properties of the implant influence the stress shielding effect. This study investigates whether the design changes from the successful ABG-I to the ABG-II stem can be verified in perioprosthetic bone remodelling using Dual-Energy X-ray Absorptiometry (DEXA).

Methods: 51 THA patients (22f, 29m, avg. age: 60.8 years) were randomised to either ABG-I or ABG-II. DEXA measurements were performed preoperatively and 10 days (baseline), 3 weeks, 3, 6, 12 and 24 months postoperatively using standard Gruen zone analysis. At the same time clinical Merle d’Aubigne (MdA) scores were measured. Changes in bone mineral density (BMD) were expressed as percentage changes from the baseline for each of the Gruen zones (R).

Results: The average MdA score (25 ABG-I, 26 ABG-II) increased from 10.3 preoperatively to 17.3 at 24 months postoperative. The improvement was higher for ABG-II (7.5) than ABG-I (6.5) but not significant (p=0.15). During the first three postoperative months the average BMD of all zones combined dropped steeply for both the ABG-I (−5.5%) and ABG-II (−4.5%, n.s.). Beyond 3 months, the overall BMD change (zones combined) continued to develop without significant difference between both implant designs (plateau and slight recovery) but the individual zones showed distinct differences. The average BMD loss in the proximal Gruen zones was much lower for ABG-II (R1: −7.9%, R7: −3.7%) than for ABG-I (R1: −9.3%, R7: −11.9%) while distally the situation was reversed with better bone preservation for the ABG-I (R3: −2.9%, R4: −1.5%, R5: −1.7%) than for the ABG-II (R3: −6.0%, R4: −2.8%, R5: −4.6%). In the mid-stem region a transitional area was identified with better bone preservation for ABG-II in Gruen zone 6 (+2.7% vs −1.4%) and for ABG-I in Gruen zone 2 (-4.9% vs 7.9%). However, the p-values (two-sided t-test) ranged from 0.05–0.35 at statistically non-significant levels.

Discussion: The steep initial bone loss for both stem designs and all Gruen zones combined indicates that during this early postoperative phase surgical trauma and reduced loading dominate the bone remodelling process and not the type of implant. The different development of proximal and distal BMD for ABG-I and II in the period thereafter demonstrates the long-term effect of implant design verifies the design improvements (less proximal stress shielding). A parallel study identified the dominant influence of preoperative BMD on BMD loss. This explains our high standard deviation and the lack of statistical significance. The study is now expanded with patients matched for preoperative BMD.

Introduction Periprosthetic bone loss is a major cause of concern in patients undergoing total hip arthroplasty (THA). In this study we monitored the periprosthetic bone loss around two different types of femoral implant to evaluate their design and investigate the relationship with the preoperative bone mineral density (BMD).

Materials and Methods Sixty patients (35 female, 25 male, mean age 63 (range, 46–75) years) undergoing THA were randomised to either the ABG or Mallory Head femoral stem. Preoperative DEXA scans were acquired of the posteroanterior (PA) and lateral lumbar spine, the contra-lateral hip and the non-dominant forearm. Postoperative DEXA scans were performed at 10 days (treated as baseline), six weeks, and 3, 6, 12, 24 and 36 months after THA.

Results A total of 50 patients (24 ABG, 26 MH) completed the study. Three months after THA there was a statistically significant BMD decrease in every Gruen zone that varied between 5.6% and 13.8% for the ABG pros-thesis and between 3.8% and 8.7% for the MH prosthesis. Subsequently, in most zones BMD reached a plateau or showed a small recovery. However, BMD continued to fall in Gruen zones 1 and 7 in ABG patients and Gruen zone 1 in MH patients. Bone loss was less in every Gruen zone in MH patients compared with ABG with the largest difference (10%, P = 0.018) in Gruen zone 7. After adjustment for multiple comparisons the relationship between peri-prosthetic bone loss and preoperative BMD was highly statistically significant for spine, hip and radius BMD.

Discussion The present study showed that statistically significant bone loss occurred with both stems during the first three months following implantation, and confirmed that prosthesis design influences peri-prosthetic bone loss. Although the greater bone loss in patients with lower spine, hip or forearm BMD may not adversely affect the outcome for patients in the short and mid-term, preservation of femoral bone stock may be important in the longer term because a low femoral BMD can adversely affect the results of a subsequent revision. Greater femoral bone stock may be particularly advantageous in younger patients in whom a future revision is more likely.