Abstract
Introduction
Wear debris generated by total hip replacements (THRs) may cause mechanical instability, inflammation, osteolysis and ultimately implant loosening, thus limiting the lifetime of such devices [1]. This has led to the development of biocompatible coatings for prostheses. Silicon nitride (SiN) coatings are highly wear resistant and any resultant wear debris are soluble, reducing the possibility of a chronic inflammatory reaction [2]. SiN wear debris produced from coatings have not been characterized in vivo. The aim of this research is to develop a sensitive method for isolating low volumes of SiN wear debris from periprosthetic tissue.
Methods
Commercial silicon nitride particles of <50nm (Sigma Aldrich) were incubated with formalin fixed sheep synovium at a volume of 0.01mm3 /g of tissue (n=3). The tissue was digested with papain (1.56mg/ml) for 6h and subsequently proteinase K (1mg/ml) overnight. Proteinase K digestion was repeated for 6h and again overnight, after which samples appeared visibly homogeneous [Figure 1]. Samples were then subjected to density gradient ultracentrifugation using sodium polytungstate (SPT) [3]. The resulting protein band was removed from the pellet of particles. Control tissue samples, to which no particles were added, were also subjected to the procedure. Particles were washed with filtered water to remove residual SPT using ultracentrifugation and filtered onto 15nm polycarbonate filters. The filtered particles were imaged by cold field emission scanning electron microscopy (CFE-SEM) and positively identified by elemental analysis before and after the isolation procedure. To validate whether the isolation method affected particle size or morphology, imaging software (imageJ) was used to determine size distributions and morphological parameters of the particles. A Kolmogorov-Smirnov test was used to statistically analyse the particle morphology.
Results
The appearance of particles was similar before and after the isolation procedure [Figure 2]. Scanning electron micrographs also demonstrated the complete removal of proteins and light impurities. Elemental analysis confirmed the identity of retrieved particles as SiN. The particle size distributions of isolated and non-isolated particles were similar [Figure 3]. Statistical analysis demonstrated that morphology in terms of roundness and aspect ratio was unchanged by the isolation procedure (P<0.05).
Discussion
Results indicate that the particle isolation method effectively isolates low volumes of SiN particles whilst retaining particle characteristics and enabling particle characterisation. The method will therefore be validated for application to additional particle materials and applied to in vivo studies of novel SiN coated prostheses in a rabbit and sheep model.
