2018 Volume 67 Issue 5 Pages 660-667
In this study, the applicability of rapid bacterial genetic testing of synovial fluid samples or intraoperative periprosthetic samples was investigated. Three different strains of Staphylococcus (methicillin-sensitive Staphylococcus aureus, MSSA; methicillin-resistant Staphylococcus aureus, MRSA; and methicillin-resistant coagulase-negative staphylococci, MR-CNS) were detected in 98 synovial fluid samples and 140 periprosthetic samples from around prosthetic joints by the PCR lateral flow method (PCR-LF), and other bacterial strains were detected by 16S ribosomal RNA (16S-rRNA) sequencing for bacterial identification. The results obtained by the PCR-LF method revealed that 24 synovial fluid samples (MSSA, 5.1%; MRSA, 9.2%; MR-CNS, 10.2%) contained bacterial DNA. Seventy-four samples with negative results of the PCR-LF method were further tested using 16S-rRNA sequencing for bacterial identification. However, a problem with detection sensitivity arose owing to Escherichia coli (E. coli) genome contamination of the enzyme solution used for amplification as well as additional bacteria present in the reagents used in the 16S-rRNA sequencing. Therefore, bacterial infection was tested using Taq polymerase from eukaryotes (E-Taq). Results showed that the direct detection of bacterial genes in samples containing bacterial cells as low as 1 × 102 CFU/mL (50% hit rate) was possible. Bacterial genes were detected and bacterial strains were identified in 19 of 74 samples by 16S-rRNA sequencing combined with the test using E-Taq. The detection rate was 1.5-fold higher by genetic testing than by the culture identification method. Notably, bacterial genes were detected in eight samples (MSSA, 3; MR-CNS, 3; MRSA, 2) that tested negative in the culture identification method and in three periprosthetic samples (MSSA, 2; MR-CNS, 1) intraoperatively collected. Therefore, contamination at the time of sample collection was suspected. In conclusion, the PCR-LF method and 16S-rRNA sequencing combined with the test using E-Taq showed satisfactory sensitivity and may therefore be clinically useful for diagnosing prosthetic joint infection.