Utility of lavage in addition to native fluid collection during fluoroscopically guided joint aspiration in infection diagnosis.

Image-guided prosthetic joint aspirations have been criticized in the literature as having poor sensitivity and specificity. Native fluid is typically analyzed for the presence of infection. Joint lavage during fluoroscopically guided aspiration of prosthetic joints is not routinely performed, and the lavage aspirate is not typically analyzed unless native fluid could not be aspirated for culture. This study aims to determine if concordance of culture results from native fluid and an additional joint lavage sample improves sensitivity and specificity in the diagnosis of prosthetic joint infection by fluoroscopically guided joint aspiration. A retrospective review of the fluoroscopically guided joint aspirations at our institution between December 2007 and December 2009 was performed. Data collected from the electronic medical record included culture results for both native fluid and lavage aspirate, histopathology results, final clinical diagnosis, as well as clinical/surgical management. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and likelihood ratio for infection were calculated for the following culture analysis scenarios: native fluid alone; lavage aspirate alone; native fluid; and lavage aspirate. Five hundred sixty-three aspirations, mostly prosthetic joints, were identified in the reviewed time period, of which 397 were sent for both native fluid and lavage aspirate analysis. Concordance between positive culture results from native fluid and joint lavage aspirate markedly increased the likelihood ratio for infection from 23.8 to 138.7 and improved specificity and PPV of fluoroscopically guided joint aspiration, with similar sensitivity and NPV to that of native fluid culture alone.