Joint aspiration is currently the reference standard test for diagnosing periprosthetic joint infection (PJI) despite the high rate of false-negative results, of which a major cause is the fastidious nature of some microorganisms. A rapid diagnostic test that detects alpha defensin (Synovasure™, Zimmer, Warsaw, IN, USA) in joint fluid can provide the diagnosis of PJI within a few minutes across the full spectrum of causative organisms (including mycobacteria and yeasts). Its performance in detecting bacterial infections is unaltered by concomitant antibiotic therapy. Few studies of Synovasure™ have been conducted by groups that were involved in designing the test, which has not been validated in France. Assessments in referral centres where complex microbiological situations are common hold considerable interest. The objective of this prospective study was to determine the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and causes of error of Synovasure™ used to diagnose periprosthetic infection in complex microbiological situations.

The rapid diagnostic test Synovasure™ has greater than 90% NPV for detecting periprosthetic infections in complex microbiological infections.

Synovasure™ was used 42 times in 39 patients between October 2015 and October 2017 in challenging microbiological situations [discordant joint aspiration results (n=20), negative cultures with clinical or laboratory evidence of infection, (n=21), and concomitant antibiotic therapy (n=1)]. Of the 39 patients, 23 had total knee prostheses, 13 total hip prostheses, and 3 total femoral prostheses. The reference standard to which the Synovasure™ results were compared was the PJI criteria set developed by the Musculoskeletal Infection Society (MSIS).

Synovasure™ was negative in 30 cases with negative joint fluid cultures (30/42, 71.4%). Of the 12 (28.6%) cases with positive Synovasure™ results, only 7 (7/12, 58.3%) had positive joint fluid cultures. According to the MSIS criteria 9 cases were infected, including 8 with positive and 1 with negative Synovasure™ results. Of the 33 cases that were not infected according to MSIS criteria, 29 had negative and 3 positive Synovasure™ results; the remaining case had a positive Synovasure™ result but was excluded when metallosis was found intra-operatively. NPV was 96.7%, PPV 72.7%, sensitivity 88.9%, and specificity 90.6%.

The high NPV of Synovasure™ suggests a role for this test in microbiologically complex situations as a new tool for ruling in and, most importantly, ruling out infection in doubtful cases.

III, prospective study of diagnostic accuracy.