Abstract
Aim
Apart from other biomarkers isolated in the synovial fluid, alpha-defensin appears to be a promising diagnostic tool to confirm a periprosthetic joint infection (PJI) in the hip or knee. The purpose of this study was to evaluate the sensitivity and specificity of an alpha defensin lateral flow (ADLF) test compared to usual standard classifications in the diagnostic management of PJI.
Method
This investigation was set up as a multicenter prospective cohort study. Synovial fluid was obtained by means of joint aspiration or intra-operative tissue biopsies. A presumptive PJI diagnosis was made according to criteria outlined by the Musculoskeletal Infection Society (MSIS), the Infectious Diseases Society of America (IDSA) and the European Bone and Joint Infection Society (EBJIS). The intention to treat by the surgeon was logged. Sensibility and specificity for the ADLF test was plotted for each aforementioned diagnostic algorithm. Spearman correlations between all scores were analyzed. Multiple logistic regression was used to determine the contribution of independent variables to the probability of PJI.
Results
Hundred thirty-six patients with a painful arthroplasty were assessed for infection and rated by the treating surgeon as potentially infected or not on the basis of clinical and laboratory information. According to the EBJIS criteria sixty-eight patients were deemed infected, fifty according to the IDSA criteria, forty-one according to the MSIS criteria and forty according to the ADLF test. However, the sensitivity of ADLF test was 87.8% for MSIS, 70% for IDSA and 55.8% for EBJIS. The specificity of ADLF test was between 94% – 97%. Good correlation was observed between synovial fluid culture and ADLF test (r = 0.73). Low to excellent correlations between the ADLF test and the EBJIS (r = 0.58), IDSA (r = 0.68), and MSIS score (r = 0.84) were observed. The surgeon's intention to treat correlated well with the MSIS score (r = 0.86), and moderately with the EBJIS (r = 0.59).
Conclusions
ADLF test sensibility was variable, but its specificity was excellent. Most of the cases, not retained by MSIS but classified by EBJIS as infected, got a negative microbiological result. Considering an accepted 20% negative microbiological result rate in PJI diagnostic, EBJIS is clearly overestimating the number of infected cases. MSIS score correlates with the surgeon intention to treat and ADLF test.
