Abstract
Introduction: The assessment of white blood cell (WBC) count and polymorphonuclear cell (PMN) percentage in synovial fluid can help in the diagnosis of acute postoperative peri-prosthetic joint infection (PJI). Their cutoff values, which would differ from those for chronic PJI, have not yet been determined in acute postoperative PJI. The aim of this study was (1) to analyse studies reporting the optimal cutoff values for WBC count and the PMN percentage in synovial fluid and (2) to determine which is the best diagnostic tool for acute postoperative PJI. Methods: We performed a systematic review (SR) of primary studies analysing WBC count and the PMN percentage for diagnosis of acute postoperative PJI. A search was performed in MEDLINE and EMBASE. We studied the risk of bias and quality assessment. We extracted data on cutoff values, sensitivity, specificity, positive and negative predictive value, area under the curve, and accuracy. We calculated the diagnosis odds ratio (DOR), performed the meta-analysis and summarized receiver operating curves (sROCs) for WBC count and the PMN percentage. Results: We included six studies. WBC count showed a DOR of 123.61 (95 % CI: 55.38-275.88), an sROC with an area under the curve (AUC) of 0.96 (SE: 0.009) and a Q index of 0.917. The PMN percentage showed a summary DOR of 18.71 (95 % CI: 11.64-30.07), an sROC with an AUC 0.88 (SE: 0.018) and a Q index of 0.812. Conclusion: We concluded that WBC count and the PMN percentage are useful tests for the diagnosis of acute PJI; WBC is the more powerful of the two. Studies centred on other synovial fluid biomarkers not yet studied could help in this diagnosis.