Abstract
Introduction
Total ankle replacement (TAR) is less successful than other joint replacements with a 77% survivorship at 10 years. Predominant indications for revision include: Insert dislocation, soft tissue impingement and pain/stiffness. Insert edge-loading may be both a product and cause of these indications and was reported to affect 22% of patients with the, now withdrawn from market, Ankle Evolutive System (AES) TAR (Transysteme, Nimes, France). Compressive forces up to seven times body weight over a relatively small contact area (∼6.0 to 9.2 cm2), in combination with multi-directional motion potentially causes significant polyethylene wear and deformation in mobile-bearing TAR designs. Direct methods of measuring component volume (e.g. pycnometer) use Archimedes' principle but cannot identify spatial changes in volume or form indicative of wear/deformation. Quantitative methods for surface analysis bridge this limitation and may advance methods for analysing the edge loading phenomena in TAR.
Aim
Determine the frequency of edge loading in a cohort of explanted total ankle replacements and compare the quantitative surface characteristics using a novel explant analysis method.
Methods
Thirty-two AES TAR devices were implanted and retrieved by the same surgeon (UK Health Research Authority approval: 09/H1307/60). Mean implantation time was 7.8 years (1.5 to 12.1 range). Pain and/or loosening were the primary indications for revision. An Alicona Infinite microscope measured the entire superior surface of each insert (10× mag; 1.76µm lateral resolution). Abbott-Firestone curves were produced per insert to quantify the deviation of the insert surface from flat. Peak material volume (Vmp), core material volume (Vmc), core void volume (Vvc) and dale void volume (Vvv) were measured. Edge loading was identified visually by a depressed area in the insert surface indicative of articulation with the edge of the tibial component. Inserts were identified as either edge-loaded or not edge-loaded and the above analyses compared.
Results
Seventeen inserts (53%) showed edge loading. Peak material volume (Vmp) was significantly increased for the edge loaded inserts 5.64 ± 5.42µm compared to the normal inserts 1.29 ± 0.954µm (Independent T-Test, P=0.005). No difference was found for the other volume parameters (Figure 2). A progressive change in insert form, beginning at the edges of the superior insert surface, was evident (Figure 1). Machining marks identified at the centre of several components supported this observation.
Discussion
Insert edge loading affected 53% of TAR explants. The volume parameters showed a statistically significant inflection of material at the inserts' edge for the affected ankles. Spatial changes to insert form progressed over time in-vivo. Machining marks at the centre of several inserts remained which indicated the deformation/wear process commenced at the periphery of the insert. Normal ranges of volume change/redistribution are not established for TAR devices and the implications of insert form change are not yet understood. However, edge-loaded components composed over half of this cohort, which reflects the conflict between design simplicity and kinematic complexity.
For any figures or tables, please contact authors directly (see Info & Metrics tab above).
