What is an acceptable sensitivity, specifi city, negative predictive value (NPV) and positive predictive value (PPV) for a diagnostic tool for periprosthetic joint infections (PJIs)?

Authors: Noam Shohat, Susan Odum
RECOMMENDATION: The validity of a diagnostic tool is traditionally measured by sensitivity, specifi city, PPV and NPV. A perfect diagnostic tool would be able to correctly classify 100% of patients with PJIs as infected and 100% of aseptic patients as non-infected. Without a perfect test available, we are left to balance between sensitivity and specifi city; increasing one would reduce the other. To reduce the rates of false positives and negatives it is extremely important to take into account the pretest probability for infection, derived from patient risk factors, clinical examination and any other examinations available at the point of assessment.
Table 1. Variety of diagnostic tools for PJI
LEVEL OF EVIDENCE: Moderate
DELEGATE VOTE: Agree: 79%, Disagree: 10%, Abstain: 11% (Super Majority, Strong Consensus)

am04