Diagnostic Sensitivity and Specificity of Synovial Biomarkers in Knee Periprosthetic Infections

Periprosthetic infections are challenging clinical situations in orthopedic surgery. Although it has been investigated by various methods over the years, there is still no clear consensus on diagnosis.

Our aim in this study is to investigate the diagnostic value of synovial biomarkers, which have gained popularity recently, in periprosthetic infections. Therefore, the diagnostic accuracy of five promising biomarkers in knee periprosthetic infections is evaluated in this study.

Synovial fluid was obtained by performing sterile puncture from the affected knees of patients with suspected prosthetic knee joint infection and meeting the inclusion criteria, and the material was divided into two aliquots. One of the samples obtained was delivered to the microbiology laboratory; gram staining, direct microscopic examination, and cell count were performed, and a culture study was carried out by incubating for five days. Eosin-methylene blue (EMB) agar, blood agar, and Sabouraud dextrose agar (SDA) were used for culture. According to the Musculoskeletal Infection Society (MSIS) criteria, patients were divided into two groups as infected and non-infected. The other sample was delivered to the medical biochemistry laboratory; levels of five biomarkers including bactericidal/permeability-increasing protein-1 (BPI), human neutrophil defensin 1-3 (HNP1-3, alpha-defensins), leukocyte elastase (LE), lactoferrin, and neutrophil gelatinase-associated lipocalin (NGAL) in synovial fluid were studied. BPI, a protein present in granules, has a high affinity for lipopolysaccharides on the bacterial membrane. HNP1-3 is a peptide found in neutrophil azurophilic granules and has a wide spectrum antimicrobial activity. LE is an enzyme deposited in azurophilic granules and has the capability of lysing the cell wall. Lactoferrin is deposited in specific granules and inhibits bacterial growth by binding iron. NGAL is similarly deposited and blocks bacterial iron metabolism. Biomarker levels were analyzed, and the diagnostic value of biomarkers was determined. Patients were also subjected to technetium-99m methylene diphosphonate (Tc-99m MDP) three-phase bone scintigraphy, and the diagnostic value of scintigraphy was assessed.

It was determined that the BPI pg/ml values of infected patients were higher than the values of non-infected patients and that there were statistically significant differences between them (p=0.002). Correlation analysis of biomarkers was done. In the receiver operating characteristic (ROC) curve analysis of the correlation of HNP1-3 with lactoferrin, 40% sensitivity and 100% specificity were detected (positive predictive value (PPV): 70%; negative predictive value (NPV): 100%). The correlation of HNP1-3 with NGAL was found to have 83.3% sensitivity and 100% specificity in the ROC curve analysis (PPV: 50%; NPV: 100%).

The findings of this study differ from those reported in the literature regarding the sensitivity, specificity, and diagnostic accuracy of biomarkers for the diagnosis of periprosthetic infections. However, it was determined in this study that the alpha-defensin-NGAL biomarker combination stands out with its high sensitivity and specificity rate in detecting periprosthetic infection and has a high diagnostic value.