Improving the accuracy of synovial fluid analysis in the diagnosis of prosthetic joint infection with simple and inexpensive biomarkers: C-reactive protein and adenosine deaminase

Bone Joint J. 2017 Mar;99-B(3):351-357. doi: 10.1302/0301-620X.99B3.BJJ-2016-0684.R1.


Aims: The aims of this study were to increase the diagnostic accuracy of the analysis of synovial fluid in the differentiation of prosthetic joint infection (PJI) by the addition of inexpensive biomarkers such as the levels of C-reactive protein (CRP), adenosine deaminase (ADA), alpha-2-macrogloblulin (α2M) and procalcitonin.

Patients and methods: Between January 2013 and December 2015, synovial fluid and removed implants were requested from 143 revision total joint arthroplasties. A total of 55 patients met inclusion criteria of the receipt of sufficient synovial fluid, tissue samples and removed implants for analysis. The diagnosis of PJI followed the definition from a recent International Consensus Meeting to create two groups of patients; septic and aseptic. Using receiver operating characteristic curves we determined the cutoff values and diagnostic accuracy for each marker.

Results: There were 23 PJIs and 32 patients with aseptic loosening. The levels of total leucocyte count, proportion of polymorphonuclear leucocytes (PMNs), CRP, ADA and α2M in the synovial fluid were all significantly higher in those with a PJI than in those with aseptic loosening. The levels of procalcitonin were comparable in the two groups. Cutoff values for the optimal performance in the diagnosis of infection were: total leucocyte count > 1463 cells/μL (sensitivity (Sens) 100%, specificity (Spec) 71.9%, positive predictive value (PPV) 71.9%, negative predictive value (NPV) 100%); proportion of PMNs > 81% (Sens 78.3%, Spec 75.0%, PPV 69.2%, NPV 82.8%); CRP > 6.7mg/L (Sens 78.3%, Spec 93.8%, PPV 90.0%, NPV 85.7%); ADA > 61U/L (Sens 78.3%, Spec 96.9%, PPV 94.7%, NPV 86.1%) and α2M > 958 mg/L (Sens 47.8%, Spec 96.9%, PPV 91.7%, NPV 72.1%). The addition of a raised level of CRP or ADA to the total leukocyte count increased the specificity: total leukocyte count > 1463 cells/μL and CRP > 6.7mg/L (Sens 78.3%, Spec 100%, PPV 100%, NPV 86.5%) or with ADA > 61U/L (Sens 78.3%, Spec 96.9%, PPV 94.7%, NPV 86.1%).

Conclusion: The total leucocyte count in the synovial fluid offers great negative predictive value in the diagnosis of PJI and the addition of more specific markers such as CRP and ADA improves the positive predictive value. Thus the addition of simple and inexpensive markers to the measurement of the leucocyte count in the synovial fluid may reduce the number of equivocal results which demand more expensive investigation. Cite this article: Bone Joint J 2017;99-B:351-7.

Keywords: Adenosine deaminase; Biomarker; C-reactive protein; Leukocyte count; Prosthetic joint infections; Synovial fluid.

MeSH terms

  • Adenosine Deaminase / analysis
  • Aged
  • Arthroplasty, Replacement, Hip
  • Arthroplasty, Replacement, Knee
  • Biomarkers / analysis
  • C-Reactive Protein / analysis
  • Calcitonin / analysis
  • Female
  • Hip Prosthesis / adverse effects*
  • Humans
  • Knee Prosthesis / adverse effects*
  • Leukocyte Count
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Preoperative Care / methods
  • Prospective Studies
  • Prosthesis Failure / etiology
  • Prosthesis-Related Infections / diagnosis*
  • Prosthesis-Related Infections / surgery
  • ROC Curve
  • Reoperation
  • Sensitivity and Specificity
  • Synovial Fluid / chemistry*
  • Synovial Fluid / cytology
  • alpha-Macroglobulins / analysis


  • A2M protein, human
  • Biomarkers
  • alpha-Macroglobulins
  • Calcitonin
  • C-Reactive Protein
  • Adenosine Deaminase