Commentary In their article, Toler et al. evaluated the diagnostic performance of a recently launched synovial fluid antigen test for the preoperative diagnosis of periprosthetic joint infection (PJI). The test is of great interest because identifying the causative microorganism prior to the surgical procedure is important in guiding surgical decision-making and starting targeted antimicrobial treatment as soon as possible. Toler et al. demonstrated a high concordance between synovial fluid culture and the antigen test. Moreover, in culture-negative synovial fluids, the test identified a microorganism in 49% of cases. However, from a clinical point of view, there are important limitations that should be taken into consideration. With regard to how the test will guide antibiotic treatment, the following shortcomings should be taken into account. A limited number of species (i.e., Staphylococcus, Enterococcus, and Candida species) are included in the test, as noted in the article, and it is important to keep in mind that the reference test used to calculate the diagnostic accuracy was a positive synovial fluid culture for the included species. A comparison with intraoperative tissue cultures was not performed, and, because the sensitivity of synovial culture is poor, an infection cannot be ruled out in the case of a negative antigen test. Therefore, empirical antibiotic treatment should still be administered in the case of a negative antigen test when a PJI is suspected. In the case of a positive test, it is important to realize that, in addition to the inclusion of a limited number of species that can be detected, the test only identifies species on a genus level. Because the antibiotic treatments for methicillin-sensitive staphylococci compared with methicillin-resistant staphylococci and for Enterococcus faecalis compared with Enterococcus faecium are different, their identification on a genus level cannot fully target antibiotic treatment. Other tests that overcome this limitation are available, such as the recently launched multiplex polymerase chain reaction (PCR) for bone and joint infections1. The BioFire Joint Infection PCR Panel provides a rapid diagnosis (i.e., within 1 hour), includes more genera and identifies pathogens on a species level, and detects resistance genes, which may better guide antimicrobial treatment1. Also, other microorganisms not detected by the PCR test may be involved. According to the literature, around 30% to 40% of PJIs are polymicrobial in nature, with an even higher incidence observed in early postoperative and chronic infections with a sinus tract2,3. Consequently, empirical antibiotic treatment cannot be easily narrowed in the case of a positive antigen test either. With regard to how the test will help us in surgical decision-making, some experts in the field have advocated a 2-stage exchange instead of 1-stage exchange for PJI caused by difficult-to-treat microorganisms such as enterococci and Candida4,5. Therefore, isolating these species prior to the surgical procedure may guide surgical decision-making. However, in the total cohort in the study by Toler et al., Candida species already grew on synovial fluid culture in 338 cases and Enterococcus species grew on culture in 465 cases. More rapid identification that the antigen provides is not needed in chronic infections, as one has time to wait for the final culture results. In the remaining preoperative samples with negative synovial fluid cultures, the antigen test detected Candida in an additional 142 cases and enterococci in an additional 188 cases. If these extra cases were to be considered to represent real infections, rather than false-positives as stated by Toler et al., only 0.3% of candidal infections and 0.4% of enterococcal infections from the total cohort of patients will have been missed when solely relying on preoperative cultures of synovial fluid. This low percentage of missed infections makes one wonder whether the number needed to test is really justified. Last but not least, Toler et al. analyze their results according to whether the preoperative diagnosis, according to the 2018 International Consensus Meeting, was positive or negative, but the inconclusive cases (7.5% of the total cohort) are not included in the final analysis, although these are the cases that are most clinically challenging. Based on the above-mentioned limitations and the way that the results are analyzed, it seems that the evaluated antigen test will not change our daily clinical practice in a large number of cases. Additional analyses, particularly comparisons with the final postoperative diagnosis and intraoperative culture results, are needed to define the role of this test in the diagnostic workup.
Read full abstract