Improved detection and treatment of prosthetic joint infection

INTRODUCTION: Although it is widely considered that aseptic loosening is the most common cause of failure of total hip replacement (THR), studies at Queen’s University Belfast indicate that routine microbiological practice underestimates the incidence of infection. By sampling directly from prostheses retrieved at the time of revision operation and adhering to strict anaerobic culture techniques the anaerobicbacterium Propionibacterium acnes is isolated as frequently as coagulase negative staphylococci. In addition non-culture detection using specific antibodies enables the detection of bacteria inculture negative samples. We now present data relating to the study of prostheses retrieved from a further 125 patients and in addition examine bone samples. METHODS: Samples were obtained from 125 patients undergoing revision operations for totalhip arthroplasty at Musgrave Park Hospital, Belfast. Upon removal from the patient, sampleswere transferred into an anaerobic jar for transport to the laboratory where they were transferred to an Anaerobic Work Station (Don Whitley) for processing. Samples were placed in pre-reducedquarter strength Ringer’s solution containing L-cysteine and then subjected to mild ultrasoundtreatment to dislodge adherent biofilm. Samples of the Ringer’s solution (sonicate) were then spread plated onto blood agars and incubated aerobically and anaerobically. Sonicate samples were also incubated with antibodies that reacted with either Staphylococcus spp or Propionibacterium acnes and FITC-conjugated secondary antibodies and then viewed using a fluorescence microscope. RESULTS: Of the 125 patients, 55 were male, 62 female and there was no gender information for 8patients. The mean age was 70 years (34 to 91); age details were absent in 12 cases. The meantime-span of the implant in situ before failure and the need for revision surgery was 12.1 years (range of 1 year to 22 years). The study confirmed our previous data that handling of samples using strict anaerobic practice resulted in the isolation of Propionibacterium acnes as frequently ascoagulase negative staphylococci. In addition, bacteria could also be cultured from bone samplesafter mild ultrasound treatment and detected by immuno-fluorescence microscopy (IFM). IFM revealed characteristic aggregates of bacteria that had been dislodged from adherent biofilm. Ninety seven percent of the patients in the study had evidence of osteolysis. DISCUSSION & CONCLUSIONS: Clearly, bacterial infection in THR is not restricted tobacteria growing in biofilm on the prosthesis surface, but may include biofilm colonisation ofthe adjacent bone. This has important implications for the potential role of bacteria in the osteolysisobserved in these patients. These data also highlight the potential importance of the bacteriumPropionibacterium acnes in the failure of total hip replacements. The potential virulence of thisbacterium is becoming increasingly clear, with evidence of variable expression of putativedeterminants of virulence amongst different types.