Scores for likelihood classification of acute coronary syndrome: What is the evidence?

Despite decades of health outcomes research, a fail-safe way to rule out acute coronary syndrome (ACS) at the time of a patient's initial presentation remains elusive. Although there may be variability in the rates at which ACS is missed at individual hospitals, the "missed ACS rate" is generally inversely proportional to the rule-in rate. Predicting the patients who will ultimately be diagnosed with ACS is particularly problematic for that subset of patients with acute chest pain but with normal initial cardiac biomarkers and a non-diagnostic electrocardiogram. This chapter will summarize the available prediction instruments that can be used for likelihood classification of ACS, including the most contemporary risk scores. How these tools address the need for intensive care and predict short-term prognosis will be explored. Data comparing these scores will also be reviewed. Lastly, a discussion of the pros, cons, and applications of several validated models to classify the presence or absence of ACS in undifferentiated ED populations will be presented.