BACKGROUND: Diagnosing a periprosthetic joint infection (PJI) is difficult and often requires a combination of objective findings. The purpose of this study is to determine whether platelets, a known acute phase reactant, would be able to further aid in the diagnosis of PJI. METHODS: A single-institution retrospective review study was performed on all revision total hip and knee arthroplasties done between 2000 and 2016 (n = 4939). PJI was defined by Musculoskeletal Infection Society criteria (n = 949). Platelet count and mean platelet volume were assessed from each patient's preoperative complete blood count. These values were then assessed as a ratio via receiver operating characteristic (ROC) curve analysis. RESULTS: The platelet count to mean platelet volume ratio for PJI patients was 33.45 compared to 25.68 for patients in the aseptic revision cohort (P < .001). ROC curve analysis demonstrates that a ratio of 31.70 has a sensitivity of 48.10 (95% confidence interval 44.9-51.4) and a specificity of 80.85 (95% confidence interval 79.6-82.1). This specificity was higher than that of both estimated sedimentation ratio (ESR) and C-reactive protein (CRP) for the same cohorts using optimal values determined via ROC curve analysis. When used in conjunction with ESR and CRP, there was a statistically significant increase in the diagnostic performance of the model used to assess PJI relative to the model that just employed ESR and CRP (P < .05). CONCLUSION: Our study demonstrates that platelets and their associated serum biomarkers are associated with PJI and warrant consideration in patients who are being evaluated for potential PJI.
BACKGROUND: Diagnosing a periprosthetic joint infection (PJI) is difficult and often requires a combination of objective findings. The purpose of this study is to determine whether platelets, a known acute phase reactant, would be able to further aid in the diagnosis of PJI. METHODS: A single-institution retrospective review study was performed on all revision total hip and knee arthroplasties done between 2000 and 2016 (n = 4939). PJI was defined by Musculoskeletal Infection Society criteria (n = 949). Platelet count and mean platelet volume were assessed from each patient's preoperative complete blood count. These values were then assessed as a ratio via receiver operating characteristic (ROC) curve analysis. RESULTS: The platelet count to mean platelet volume ratio for PJI patients was 33.45 compared to 25.68 for patients in the aseptic revision cohort (P < .001). ROC curve analysis demonstrates that a ratio of 31.70 has a sensitivity of 48.10 (95% confidence interval 44.9-51.4) and a specificity of 80.85 (95% confidence interval 79.6-82.1). This specificity was higher than that of both estimated sedimentation ratio (ESR) and C-reactive protein (CRP) for the same cohorts using optimal values determined via ROC curve analysis. When used in conjunction with ESR and CRP, there was a statistically significant increase in the diagnostic performance of the model used to assess PJI relative to the model that just employed ESR and CRP (P < .05). CONCLUSION: Our study demonstrates that platelets and their associated serum biomarkers are associated with PJI and warrant consideration in patients who are being evaluated for potential PJI.