Abstract
INTRODUCTION
Although simulation studies have shown superior wear properties of metal-on-metal articulations, increased concern exists regarding the excess in-vivo wear of a small number of Metal-on-Metal-Hip-Resurfacing (MoMHRA) implants. Serum ion levels of Chromium (Cr) and Cobalt (Co) are surrogate markers of wear. Risk factors associated with increased wear include female gender, small components, dysplasia, cup orientation outside safe zone and femoral head downsize during surgery with an associated decrease in Head-Neck-Ratio (HNR). However, these factors are interlinked. This study aims to identify the factors that are most important for subsequent wear of MoMHRA, by performing a multivariate analysis.
METHODS
206 patients (124M: 82F) with unilateral MoMHRA were included in this study. The average follow up was 3.3 years. All patients had Cr/Co levels measured at follow up. Inclination and anteversion of each cup were measured using EBRA. Cups were analysed as being within or outside the previously defined optimum-zone. HNR measurements were made from pre-operative (HNRpre) and post-operative (HNRpost) radiographs. The immediate changes in HNR (downsize/upsize of femoral head) as a result of the operation were expressed as:
HNRprepost=HNRpost–HNRpre
Multivariate linear regression modelling was used to explore the association between measures of ions with the following predictor variables (gender, age, diagnosis, femoral component size, orientation of the acetabular component, head/neck ratio and position of femoral stem). Analyses were carried out separately for each outcome (Cr and Co). Classification and Regression Tree (CART) models were fitted as a complimentary approach to regression modelling.
RESULTS
Articular surface downsize followed by cup orientation within/outside optimum zone, followed by cup anteversion followed by gender were the strongest predictors of ion levels. A percentage decrease in HNRartpost, predicted an increase in Cr ion level by 5% (and 6% for Co). If the cup was within the optimum zone this was associated with decreased levels of Cr and Co ions. As acetabular component anteversion increased, levels of ions increased. Gender had a strong effect on ion levels. Adjusting for other variables, the effect of gender was attenuated due to a confounding effect of component size and the amount of femoral head downsize.
Predictors identified as important in regression analyses were similar to those produced in the CART model, where the highest levels of ions were seen in patients with a percentage decrease in pre-operative HNR −11.3.
DISCUSSION
This analysis shows that (surgical) factors, such as amount of femoral head downsize at operation and whether the cup was orientated within the optimum zone, explained the majority of the variability in ion levels in MoMHRA.
