gms | German Medical Science

Objectives: Joint aspiration is the standard pre- and intra-operative procedure to diagnose or exclude periprosthetic joint infection (PJI). Culture of joint aspirate has limited sensitivity, especially for patients with prior antibiotic therapy. Polymerase chain reaction (PCR) is a non-culture-based molecular method. We evaluated a novel, fully automated, multiplex PCR for detection of common PJI pathogens and their genotypical susceptibility. In addition, PCR may detect non-viable bacteria in patients previously treated with antibiotics and may detect low-grade agents that were not detected by aspirate culture. To assess the diagnostic accuracy of multiplex PCR of joint aspirate in comparison to current standard diagnostic methods for detection of hip and knee PJI.

Methods: In this prospective study, joint aspirates were obtained pre- and intra-operatively in all patients with suspected PJI of the hip or knee from October 2014 to April 2015 in our institution. Diagnosis of PJI was established when at least one of following criteria applied: macroscopic purulence, presence of sinus tract, positive cytology of joint aspirate (>2000 leukocytes/µl or >80% granulocytes), acute inflammation in periprosthetic tissue, positive culture. Joint aspirates were analyzed by multiplex PCR Unyvero (Curetis, Holzgerlingen). Sensitivity, specificity and accuracy of each test were determined and McNemar's Chi squared test was employed to find significant differences between the different diagnostic methods.

Results and Conclusion: 56 patients were included (33 knee and 23 hip prosthesis). The median age (range) was 72 (42-87) years; 22 (39%) were males. Diagnostic results are detailed in Table 1. The sensitivity of PCR (61%) was similar to the one of joint aspirate culture (51%), whereas the specificity was same high (100%). The PCR detected coagulase-negative staphylococci (n=8), Staphylococcus aureus (n=3), Escherichia coli (n=4), streptococci (n=5), Propionibacterium acnes (n=3) and others (n=3). The processing time for PCR was 5 hours, whereas cultures required a median of 48 hours (range 1-14 days).

Non-culture based methods (synovial aspirate cell count/differential & intra-operative histology) obtained the highest sensitivity and accuracy but do not identify the microbiological agent. The PCR provided comparable sensitivity and specificity, but faster results (<5 hours versus 48 hours), including their antibiotic susceptibility. With improvement of its sensitivity, multiplex PCR may complement or replace cultures of joint aspirates for the preoperative diagnosis of PJI.