The diagnosis of periprosthetic joint infection is challenging. Notably, low-grade infections with only subtle clinical signs of infection are difficult to discriminate from aseptic prosthetic failure and are frequently missed using insensitive definition criteria proposed by different societies. Routine diagnostic methods include preoperative joint aspiration with determination of leukocyte count and culture of synovial fluid, periprosthetic tissue sampling for histopathological and microbiological analysis, and sonication of the explanted prosthesis. Diagnostic approaches target either the causative pathogen or the inflammatory host response directed against the pathogen. Whereas all standard diagnostics are fairly specific, they lack sensitivity. Therefore, huge efforts have been made to find a novel discriminative test. We investigated three novel and culture-independent diagnostic tests in prospective cohorts of patients with failed prostheses. Sensitive institutional definition criteria were used to also capture low-grade infections. The first novel biomarker was alpha-defensin, an anti-microbial peptide, which is released by neutrophils in response to the presence of bacterial pathogens. We evaluated the qualitative bedside lateral flow test in synovial fluid and compared it to leukocyte count as standard method. Showing a high specificity (>95%), but a low sensitivity (54%-84% depending on used definition criteria), the biomarker might be useful as confirmatory test, but not as a screening method. In a second study, similar performance of the qualitative bedside lateral flow test compared to the quantitative laboratory test (ELISA; enzyme-linked immunosorbent assay) was demonstrated. The second biomarker was d-lactate, a product of bacterial metabolism, of which the concentration in synovial fluid was measured by spectrophotometry. Compared to leukocyte count, d-lactate showed similar performance for diagnosis of PJI. Advantages of D-lactate test are requirement of low synovial fluid volume, short turnaround time and low cost. In a second study, high sensitivity of d-lactate was shown independently of the used definition criteria for PJI. Furthermore, a correlation between d-lactate concentration and virulence of pathogen was shown. The third test was a multiplex PCR test kit specifically developed for periprosthetic joint infections, which was tested in synovial fluid and sonication fluid. Both studies revealed similar performance of culture and multiplex PCR. However, multiplex PCR was superior for detection of low-virulent bacteria compared to culture in synovial fluid. Furthermore, we compared phenotypic (culture) and genotypic (PCR) resistance testing in synovial and sonication fluid samples of patients with implant-associated infections. The overall sensitivity for detection of antimicrobial resistance by mPCR was low. In conclusion, the prospective evaluation of three novel tests in patients with prosthetic failure showed no novel test to be superior compared to standard diagnostic tests. However, they might be used complementary to standard diagnostic methods in specific situations. Furhtermore, D- lactate is a promising marker and further research should focus on alternative determination methods to improve the specificity.
