Purpose: Femoral neck narrowing (FNN) has been reported after metal-on-metal hip resurfacing (HR). It is significant (> 10%) in a number of cases (from 0 up to 27.6%). Its origin remains unclear, but bone remodelling, impingement, head necrosis and osteolysis have been incriminated. The aims of this study were to assess these issues and describe their consequences in a prospective series with a minimum follow-up of five years.

Method: Fifty-seven HRs in 53 patients (30 men, 23 women, average age 49.2±8.4 years) were included prospectively with clinical (WOMAC, UCLA activity score) and radiological evaluation at one, two and five years. All patients received the Durom™ resurfacing system (Zimmer, Warsaw, IN, USA), with cementless acetabular cup and cemented femoral implant. All cases were undertaken via a posterior approach. Femoral and acetabular implant positioning was assessed. The neck-to-head prosthesis (N/H) ratio was calculated at the junction of the neck with the femoral component and at mid-distance between the neck junction and the inter-trochanteric line (N1/2/H) on anterior-posterior view. Ion concentrations (chromium, cobalt and titanium) were measured at 12 months. We considered p< 0.05 as the significance level.

Results: The N/H ratio decreased significantly at one, two and five years in comparison to the postoperative data (p< 0.01 for all parameters) and N1/2/H declined significantly only at one and two years (p=0.003 and p=0.03, respectively). There was no difference in the N/H ratio or N1/2/H between two and five years. We encountered no deleterious consequences of FNN on clinical outcome, and no significant relationship with cup positioning, gender, body mass index or level of activity. Femoral positioning in valgus was associated with a decrease in N1/2/H at one and two years (p=0.02), whereas the N/ H ratio tended to be lower when cobalt concentration was elevated (p=0.08). Significant FNN was observed in two cases at two years (−12.9% and – 11.1%) with a localized and progressive femoral anterior-superior notch absent on immediate postoperative X-rays. At five years, we noted three other cases with circumferential FNN, limited at the junction neck-cup area (average narrowing around – 20% between two and five years). One of these cases presented a femoral stem fracture. Osteonecrosis was confirmed during surgical revision.

Conclusion: In the current group, FNN was seen infrequently up to five years after surgery (9%). Mechanically-induced remodelling should be differentiated from overall FNN which may be due to femoral head necrosis. In this case, revision could be proposed before implant failure or femoral loosening. Impingement causes very early and localized FNN at the upper part of the neck; for these patients, simple observation should be the rule, all the more since they are usually pain-free and rarely disabled.