Introduction – Polyethylene (PE), or better Ultra High Molecular Weight Polyethylene (UHMWPE) wear was demonstrated to be the main cause of Total Knee Replacement failure during the ‘90ties years. Wear, that occurred during the in vivo service, was related to the implant biomechanics, both the prosthetic design (constrained - non constrained, PCL sacrifice…) and the implant technique with rotational and alignment defects. In all these studies, retrieved PE inserts and wear particles were supposed to be UHMWPE, with the same chemical and physical characteristics of the original certified polymer. Unfortunately, degradation of UHMWPE, that is the modification of the chemical and physical structure, may occur during the preparation of prosthetic components; in particular, gamma irradiation in air is responsible for superficial and deep, unpredictable, inhomogeneous oxidative degradation of the polymeric biomaterial (1–6). Therefore, new PE components sterilised by in air gamma irradiation and ready for implant can be supposed to be UHMWPE, but they could not be. Sterilisation with ethylene oxide (EtO) does not modify the chemical and physical properties of the original PE. Furthermore, during the service in vivo cholesterol and other components of the synovial fluid diffuse in the PE components and modify the mechanical properties of the polymer (7).
Aim of the study - To characterise new PE components (hereafter called PEs in this paper) ready for implants and retrieved PEs obtained from failed total knee replacements in order to evaluate the wear, oxidation level and, in the retrieved ones, the diffusion products after service in vivo. Only after this characterisation some mechanical considerations and therefore wear in vivo could be discussed.
Materials - 24 new and 75 retrieved PEs were analysed. New PEs were produced by 9 different firms, 18 were sterilised by gamma irradiation in air, 1 in inert atmosphere or in vacuum, 5 by EtO. Surgical revisions were performed after an average time of 5 years (min 3 months, max 15 years) because of aseptic loosening (51 cases), septic failure (16 cases), PE severe wear (4 cases), other causes (4 cases). The retrieved PEs had been produced by 15 different firms; 74 were sterilised by gamma irradiation, while only 1 by EtO. The mean age at revision was 70 years (range 57–82 years).
Methods – At the surgical revision, PEs were photographed; wear area score according to Collier and wear severity score according to Plante-Bordeneuve and Freeman were evaluated. Prior to the analyses, PEs were stored in the dark in formaldehyde 4%. New and retrieved PEs were cut perpendicularly to the articular surface. A series of slices of controlled thickness (from 100 to 300 microns) were recovered from the cross-section using a Poly Cuts Microtome (Reichert-Jung) at 20 mms−1 in air at room temperature. A FTIR Microscope (Perkin Elmer System 2000) equipped with a x-y motorised micropositioning stage was used to identify and map the distribution and level of oxidation. Identification of oxidised species was carried out by derivatisation and IR analysis. Soxhlet extraction in boiling cyclohexane for 20 hours was performed to extract low molecular weight substances which diffused into PEs.
Results – All new PEs sterilised by gamma irradiation in air presented surface and bulk oxidation, variable in severity and distribution. Wear of retrieved PEs sterilised by gamma irradiation in air was extended for more than 50% of the articular surface (score 3) in 60% of cases and was severe (score 5–8) in 47% of them. In most of the gamma irradiated in air retrieved plateaux a “crown zone” at a depth of 1–2 mm from the surface was observed. This zone has been found to correspond to the maximum of the oxidation profile, measured by FTIR mapping. Diffusion of cholesterol and its esters with fatty acids has been observed in many of the samples, in variable amount depending both on the clinical situation of the patient and on the implant time.
Conclusions – These results show that significant modifications of the physical properties of UHMWPE can be introduced by the sterilisation treatment. The chain reactions that follow gamma ray sterilisation and lead to oxidation tends to decrease the molecular weight of UHMWPE. The presence of the subsurface “crown zone” is the macroscopic evidence of an extremely high level of oxidation, responsible for delamination and wear of the tibial plateaux. Critical mechanical phenomena can be related with oxidation due to gamma irradiation and in vivo degradation. Diffusion in the PEs of cholesterol and other synovial fluid components may affect the mechanical resistance in vivo. All these results emphasise that discussion about mechanical behaviour in vivo of different prostheses, particularly for total knee replacement where the biomechanics is complicate, must be proceeded by accurate control of physicochemical properties of the ready-to-implant prosthesis and of the retrieved components. New tests must be introduced to control the integrity of the ready-to-implant components, besides that of the row material. Furthermore, suitable in vitro tests might give a prediction on the effects of diffusion on the material performances.
