Duke criteria
The Duke criteria are a structured checklist that doctors use to decide how likely a patient is to have infective endocarditis. Because no single test is perfect, the criteria combine several lines of evidence into a scoring scheme, rather like building a legal case from many pieces of circumstantial evidence rather than one confession.
The evidence is sorted into major and minor criteria. Major criteria include persistently positive blood cultures with typical organisms and imaging evidence of endocardial involvement such as a vegetation on echocardiography. Minor criteria include predisposing heart conditions or injection drug use, fever, embolic or vascular phenomena, immune phenomena, and suggestive but not fully qualifying microbiology.
Combining these, a case is classified as definite, possible, or rejected endocarditis. The original criteria have been updated over time (the modified Duke criteria, and more recent revisions) to incorporate newer imaging such as cardiac PET and a broader range of organisms. They are a clinical aid, not a substitute for judgement, and a strong clinical suspicion can still warrant treatment even when formal criteria fall short.
The criteria are named after Duke University, where they were first proposed in the 1990s, and they have been refined several times since.