Pre-operative diagnosis of infection in total knee arthroplasty: an algorithm.

The aim of our study was to construct an algorithm for the pre-operative diagnosis of infection in total knee arthroplasty. We analyzed the currently used parameters in a consecutive series of 31 patients with failed implants. An outcome of at least 2 years was prospectively considered to validate our algorithm. Patient history, imaging, laboratory studies, histology, pre- and intra-operative cultures were considered. The optimal cutoffs of the inflammation tests for diagnosing infection were determined by constructing the receiver operating-characteristic curves. Sensitivity, specificity and accuracy of these tests as infection markers were determined. The combination of at least two tests with values higher than the cutoffs is reliable for predicting the infection. Scintigraphy, needle-aspirate cell count and culture can integrate the pre-operative evaluation. Doubtful cases can be clarified by microbiological and histological analyses. As a result an algorithm helpful to identify the cause of loosening has been developed. In our opinion, adherence to this algorithm could contribute to preoperatively define a rational surgical and antibiotic treatment strategy.