Abstract
Aim
The incidence of hematogenous periprosthetic joint infections (hPJI) is unknown and the cases probably largely underreported. Unrecognized and untreated primary infectious foci may cause continuous bacteremia, further spread of microorganisms and thus treatment failure or relapse of infection. This study aimed at improving knowledge about primary foci and microbiological characteristics of this entity to establish preventive measures and improve diagnostic and therapeutic strategies to counteract hPJI.
Method
We retrospectively analysed all consecutive patients with hPJI, who were treated at our institution from January 2010 until December 2016. Diagnosis of PJI was established if 1 of the following criteria applied:(i) macroscopic purulence, (ii) presence of sinus tract, (iii) positive cytology of joint aspirate (>2000 leukocytes/μl or >70% granulocytes), (iv) significant microbial growth in synovial fluid, periprosthetic tissue or sonication culture of retrieved prosthesis components, (v) positive histopathology. PJI was classified as hematogenous if the following criteria were fulfilled additionally: (1) onset of symptoms more than 1 month after arthroplasty AND (2) i) isolation of the same organism in blood cultures OR ii) evidence of a distant infectious focus consistent with the pathogen.
Results
A total of 70 episodes of hPJI were included. Median age was 74 years (32–89 years), 36 were women and 29 men. Sites of PJI included 39 knees, 29 hips, one shoulder and one elbow joint. The pathogen was identified in 99% (n=69), the majority of episodes was monomicrobial (n=64, 91%). Blood cultures were collected in 39 cases (56%) and identified the pathogen in 67% (n=26). Isolated pathogens were Staphylococcus aureus (n=29), Streptococcus spp. (n=20) and Enterococcus faecalis (n=12), coagulase-negative staphylococci (n=6) and gram-negative bacilli (n=5). In 55% the primary focus was identified and included an intravascular (endocarditis, endoplastitis, thrombophlebitis; n=15), urogenitary (n=8), dental (n=6), gastrointestinal (n=5) and osteoarticular (n=2) and skin and soft tissue origin (n=1). The primary focus could not be identified in 29 cases (41%), primarily due to underuse of diagnostic workup.
Conclusions
Causative agents were identified in the vast majority of hPJI with a predominance (75%) of high virulent microorganisms such as staphylococci, streptococci and gram-negative bacilli. Our results highlight the importance of a meticulous diagnostic workup including collection of blood cultures and performance of echocardiography in hematogenous PJI in order to cure the infection and prevent relapse. Awareness must be raised with regard to every prosthesis being endangered by hematogenous seeding from a distant infectious focus during the entire indwelling time.
