Laboratory tests on total joint replacement prostheses.

Total joint replacement is becoming a clinical commonplace. The hip is now often replaced, the knee and finger joints on a smaller but increasing scale, and widening interest is being shown in designs of implants for the total replacement of the elbow and shoulder joints. The largest experience (considering both numbers of patients and lengths of service) has been obtained with the hip, in which it is now known that more than one design will give at least eight years’ service and that, supposing extrapolation of the rates of wear observed in clinical use to be permissible, at least twenty years of working life should often be obtainable. Whether such extrapolation is permissible is a separate question to which attention is being devoted. Nevertheless, on the assumption that prostheses of present and foreseeable designs will not wear out in less than the time for which many patients will require them, it is wise to consider the other factors which may limit the useful life of a total replacement implant. On the basis of recorded clinical experience (Girzadas, Geens, Clayton and Leidholt 1968; Kitridou, Schumacher, Sbarbaro and Hollander 1969; Arden, Taylor and Ansell 1970; Charnley 1970; Stinchfield and White 1971 ; Arden and Ansell 1971 ; Patterson and Selby Brown 1972; Safi and Schreiber 1972; Wilson, Amstutz, Czerniecki, Salvati and Mendes 1972) and theoretical possibilities, these factors may be listed as infection, trauma, loosening and biological damage from wear or corrosion products. The prevention of infection is primarily a matter of surgical technique and facilities, and this topic will not be considered in detail in this paper. It may however be noted that the design of an implant can theoretically influence the development and consequences of infection by making insertion possible with minimal dissection and the creation of the smallest possible implant-tissue interface, and by making removal simple and not mutilating. Such factors contributed to the design of a total replacement knee by two of the authors (Freeman, Swanson and Zahir 1972; Swanson and Freeman 1972). Similarly, the occurrence of trauma is not a matter for laboratory testing, although by the choice of materials and the mechanical design of an implant it is possible to limit the damage that might be caused by incidental trauma. Loosening must be assumed to be influenced by, amongst other factors, the magnitude of the frictional forces and moments in the prosthetic joint although other factors may well be relevant. Wilson and Scales (1970) discussed the loosening of femoral head components of total hip replacements and Andersson, Freeman and Swanson (1 972) the loosening of prosthetic acetabula. The possibility of biological damage to the tissues obviously depends on the nature and quantity of the wear or corrosion products, which in turn will depend on the materials used for the bearing and the mechanical design of the implant. The tests described in this paper were designed to investigate the performance of a selection of currently used prosthetic hips and knees, and also some combinations of materials not yet in clinical use, particularly with respect to friction and the wear and corrosion products. When these tests were started, published measurements of frictional moments had given only relative and not absolute values (Scales, Duff-Barclay and Burrows 1965). Also, whereas the