**3.1 Acquisition of infection**

Acquisition of prosthetic joint infection occurs by two mechanisms: direct inoculation and haematogenous seeding. Direct inoculation of the prosthesis may occur at the time of implantation or with manipulation of the arthroplasty and is thought to be the predominant mechanism of infection. In a study by Southwood et al the 50% infective dose (ID50) of *Staphylococcus aureus* required to induce infection with direct inoculation of the prosthesis was just 50 organisms. This compared to an intravenous inoculum dose of 100 000 organisms at the time of operation for bacteraemic seeding and infection of the prosthesis to occur. Southwood also demonstrated that three weeks after implantation of the prosthesis, the likelihood of bacteraemic seeding of the prosthesis was significantly reduced. In fact, in the rabbit model, the inoculum of intravenous bacteria required was near to the lethal dose(Southwood, et al. 1985). Nevertheless, haematogenous seeding remains an important cause of arthroplasty infections and it has been reported that up to 34% of patients with prosthetic joints in-situ developed deep infection of that prosthesis following an intercurrent episode of *Staphylococcus aureus* bacteraemia (Murdoch, et al. 2001).

Whilst theoretically distinct, clinically there is significant overlap between both mechanisms of infection. The simplified view is that infection resulting from inoculation occurs within the first year of implantation whilst haematogenous infections occur later. However the clinical presentation of prosthetic joint infections acquired during the original operation may be much more delayed, particularly with low virulence organisms such as coagulase negative staphylococcus species (Steckelberg & Osmon 2000). Furthermore, up to 50% of suspected prosthetic joint infections of haematogenous origin present within the first two years (Deacon, et al. 1996). However it is important to note that distinguishing between whether an episode of bacteraemia led to haematogenous seeding of a prosthetic joint or whether the primary source of the bacteraemia was a subclinical prosthetic joint infection can be problematic.
