Abstract
Aim
Several options to standardize the definition of periprosthetic joint infection (PJI) have been created including the 2013 Musculoskeletal Infection Society (MSIS), 2018 Intentional Consensus Meeting (ICM), and the 2019 proposed European Bone and Joint Infection Society (EBJIS) criteria. Synovial fluid biomarkers have been investigated in an effort to simplify and improve the diagnosis of PJI. The aim of this study was to test the sensitivity, specificity, positive, and negative predicted values (PPV and NPV, respectively) of a calprotectin point of care (POC) test for diagnosing PJI in revision total knee arthroplasty (TKA) patients comparing different sets of criteria (2013 MSIS, 2018 ICM, and 2019 EBJIS criteria) used to define patients as with or without infection.
Method
From October 2018 to January 2020 and under IRB approval 123 intraoperative samples of synovial fluid were prospectively collected at two academic hospitals in the same institution from revision TKA patients. All patients underwent standard clinical and laboratory evaluation for PJI at our institution, allowing for categorization using the 3 criteria. Patients were adjudicated by 2 blinded and independent reviewers for the 3 sets of criteria. The 3 criteria agreed 91.8% of the time. Four likely cases by the 2019 proposed EBJIS were considered unlikely and 1 inconclusive case by the 2018 ICM was considered not infected for the purposes of analysis. Calprotectin POC testing followed manufacturer's instructions using a threshold of >50 mg/L to indicate PJI. Sensitivities, specificities, PPV, NPV, and areas under the curve (AUC) were calculated for the 3 sets of criteria.
Results
Using 2013 MSIS criteria the calprotectin POC test demonstrated a sensitivity, specificity, PPV, NPV AUC of 98.1%, 95.7%, 94.5%, 98.5%, and 0.969, respectively. Using 2018 ICM the POC test demonstrated a sensitivity, specificity, PPV, NPV and (AUC) of 98.2%, 98.5%, 98.2%, 98.5%, and 0.984, respectively. Using the 2019 proposed EBJIS criteria the POC test demonstrated a sensitivity, specificity, PPV, NPV and area under the curve (AUC) of 93.2%, 100.0%, 100.0%, 94.2%, and 0.966, respectively.
Conclusions
The calprotectin lateral flow POC test has an excellent sensitivity and specificity regardless of the set of criteria used to define PJI. These results are promising and suggest that the calprotectin lateral flow test may be used as a rule out test in a cost-conscious health care model or when conventional diagnostic tools may not be available. Further investigations of the calprotectin PCO test must be completed to validate these results.
