Abstract
Background This prospective study assessed the performance of a commercial multiplex polymerase chain reaction (PCR) panel in diagnosing periprosthetic joint infection (PJI). Methods Synovial fluid samples from hip and knee arthroplasties, submitted to a clinical diagnostic laboratory for PJI workup, were tested with multiplex PCR to identify a specific panel of organisms. Clinical biomarker test results and microbiological culture were used to classify samples as 'not infected', 'inconclusive', and 'infected' using a modified 2018 International Consensus Meeting (ICM) criteria for PJI. The purpose of this study was to evaluate the specificity, the on- and off-panel sensitivity, and the clinical utility of the multiplex PCR panel for diagnosing PJI. Results The multiplex PCR panel demonstrated 100% specificity (55/55 samples; 95% CI: 94%-100%) with no false-positive cases. Sensitivity was 96% (95% CI: 89%-99%) for the infected samples (85/89 samples), yielding microorganisms included as targets in the PCR panel. For all culture-positive samples classified as infected, including off-panel organisms, the sensitivity of the multiplex PCR panel for pathogen identification was 58% (87/150 samples; 95% CI: 50%-66%). The panel identified microorganisms in 18% (9/50) of the culture-negative samples classified as infected and in 2% (1/50) of culture-negative samples classified as inconclusive. Conclusion The multiplex PCR panel assessed in this study demonstrated excellent specificity, with a negligible false-positive rate, yet still identified organisms in the setting of culture-negative infection. Sensitivity for organisms included on the multiplex panel was also excellent. Clinicians must be aware that the panel will yield negative results in the setting of organisms which are not on the panel, such as Staphylococcus epidermidis.