Abstract
Background
Sequentially annealed, highly crosslinked polyethylene (HXLPE) has been used clinically in total knee arthroplasty (TKA) for over a decade[1]. However, little is known about the reasons for HXLPE revision, its surface damage mechanisms, or its in vivo oxidative stability relative to conventional polyethylene. We asked whether retrieved sequentially annealed HLXPE tibial inserts exhibited: (1) similar reasons for revision; (2) enhanced resistance to surface damage; and (3) enhanced oxidative stability, when compared with tibial inserts fabricated from conventional gamma inert sterilized polyethylene (control).
Methods
Four hundred and fifty-six revised tibial inserts in two cohorts (sequentially annealed and conventional UHMWPE control) were collected in a multicenter retrieval program between 2000 and 2016. We controlled for implantation time between the two cohorts by excluding tibial inserts with a greater implantation time than the longest term sequentially annealed retrieval (9.5 years). The mean implantation time (± standard deviation) for the sequentially annealed components was 1.9 ± 1.7 years, and for the control inserts, 3.4 ± 2.7 years (Figure 1). Reasons for HXLPE revision were assessed based on medical records, radiographs, and examinations of the retrieved components. Surface damage mechanisms were assessed using the Hood method[2]. Oxidation was measured at the bearing surface, the backside surface, the anterior and posterior faces, as well as the post (when available) using FTIR (ASTM F2102). Surface damage and oxidation analyses were available for 338 of the components. We used nonparametric statistical testing to analyze for differences in oxidation and surface damage when adjusting for polyethylene formulation as a function of implantation time.
Results
The tibial inserts in both cohorts were revised most frequently for loosening, infection, and instability. Instability was observed more frequently in inserts without a stabilizing post. In both cohorts, the most commonly observed surface damage mechanisms were burnishing, pitting, and scratching. Delamination was rare and only observed in 2 sequentially annealed inserts and 7 inserts in the control cohort. We observed six cases of posterior condyle fracture, which was always associated with instability (Figure 2). 5/6 of the fracture cases did not have a stabilizing post. Oxidation indices of the sequentially annealed inserts were, on average, low (ASTM oxidation index < 1) and not significantly different than the control inserts on the bearing surface and anterior/posterior face (Figure 3).
Discussion
The findings of this study document the reasons for revision, surface damage mechanisms, and oxidative behavior of sequentially annealed HXLPE for TKA. We observed evidence of low in vivo oxidation in both retrieved sequentially annealed HXLPE and control tibial inserts. We found no association between the levels of oxidation and clinical performance of the HXLPE tibial components. However, because of the short-term follow-up, analysis of longer-term retrievals may be appropriate.