Prosthetic joint infection is one of the most challenging complications of joint alloplasty and the diagnosis remains difficult. The aim of the study was to investigate the bacterial flora in surgical samples from 22 prosthetic patients using a panel of culture-independent molecular methods including broad range 16S rRNA gene PCR, cloning, sequencing, phylogeny, quantitative PCR (qPCR), and fluorescence in situ hybridization (FISH). Concomitant samples were cultured by standard methods.

Molecular methods detected presence of bacteria in samples from 12 of 22 patients. Using clone libraries a total of 40 different bacterial species were identified including known pathogens and species not previously described in association with prosthetic joint infections. The predominant species were Propionibacterium acnes and Staphylococcus epidermidis; polymicrobial infections were found in 9 patients. Culture-based methods showed bacterial growth in 8 cases with the predominant species being S. epidermidis. Neither anaerobic bacteria (including P. acnes) nor any of the species not previously described in implant infections were isolated. Additionally, 7 of the 8 culture positive cases were monomicrobial. Overall, the results of culture-based and molecular methods showed concordance in 11 cases (hereof 9 negative by both methods) and discrepancy in 6 cases. In the remaining 5 cases, culture-based methods identified only one species or a group of bacteria (e.g., coagulase negative staphylococci or coryneform rods), while culture-independent molecular methods were able to detect several distinct bacterial species including a species within the group identified by culture. A qPCR assay was developed to assess the abundance of Propionibacterium while S. aureus was quantified by a published S. aureus qPCR assay. These quantifications confirmed the findings from the clone library approach and showed the potential of qPCR for fast detection of bacteria in orthopedic samples. Additionally, both single cells and microcolonies were visualized using FISH and confocal scanning laser microscopy.

In conclusion, the molecular methods detected a more diverse bacterial flora in prosthetic joint infections than revealed by standard culture-based methods, and polymicrobial infections were more frequently observed. The pathogenesis of these microorganisms and their role in implant-associated infections needs to be determined.