Abstract
Background
Currently, the gold standard for the microbiological diagnosis remains the culturing of preoperative aspirated joint fluid and intraoperative periprosthetic tissue samples, which give false negative results in about 7 % of cases. Lytic bacteriophages are viruses that specifically infect and lyse bacteria within their replication cycle.
Aim
The aim of our study was to explore possibilities for the use of bacteriophage K for the detection of live Staphylococcus spp. bacteria in sonicate fluid of infected prosthetic joints, to possibly contribute to the development of a faster, more sensitive, specific and at the same time economical and handy method for the establishment of the right diagnosis.
Material and methods
Sonicate fluid samples obtained from 104 patients with revision arthroplasty were analysed. After the optimisation two indirect phage-based methods were used: a) bioluminescence detection of bacterial intracellular ATP released by bacteriophage K mediated lysis and b) q-PCR with primers specific for bacteriophage K DNA. The results were compared with classical microbiological cultivation methods.
Results
With both methods the analysis of sonicate fluid and the analysis of its over-night culture achieved 100 % specificity and predictive value, as there were no false positive results. The sensitivity of the methods was lower when analysing sonicate fluid samples directly, without cultivation. The sensitivity of qPCR detection was higher (81.25 %) compared to the sensitivity of ATP detection (62.5 %) in sonicate fluid directly as a result of 3 false negative results with the qPCR method compared to 6 false negatives with the ATP detection method. The sensitivity of the methods was significantly improved (to 94.12 %) with overnight cultivation of sonicate fluid samples prior to analysis, with no difference in detection between the methods. With both methods, with pre-cultivation of sonicate fluid samples, only one of the tested samples resulted in a false negative result. However, the same sample was negative even when tested with standard microbiological methods. In this patient, only the microbiological cultivation of the periprosthetic tissue sample was positive. The bioluminescence method took 3h with a limit of detection (LOD) in the bacterial concentration range of 103 CFU/mL. The method with qPCR took 4h and had a LOD of 102 CFU/mL.
Conclusion
Detection of staphylococci within sonicate fluid with bacteriophage K based methods is a rapid, sensitive and specific approach.
