Patients with chest pain and symptoms of acute coronary syndromes (ACS) account for over 600,000 emergency department (ED) visits annually in Canada. 85% of these patients do not have an ACS, and most are discharged from the ED after a thorough evaluation. However, a large proportion of these patients (approximately 200,000 annually) are referred for outpatient objective cardiac testing (exercise stress tests, myocardial perfusion scans, coronary CT angiography) after ED discharge, even though their short-term risk of major adverse cardiac events such as death, new myocardial infarction or need for revascularization is very small.

This overuse of outpatient consultation and testing is attributable, in part, to a lack of accurate risk prediction tools to distinguish patients at very low risk (who do not require additional testing) from higher-risk patients (who may benefit from additional testing). Existing risk prediction tools are very sensitive for short-term major adverse cardiac events (MACE), but they only classify about one-third of patients as low-risk, implying that two-thirds of patients should undergo additional testing in spite of the low risk of short-term adverse events.The poor classification performance of existing risk prediction tools stems from several factors:

1. They were derived in the wrong population, and sought to answer the wrong question. Some were derived in stable outpatients to predict the risk of coronary artery disease. Others were derived in undifferentiated ED chest pain patients in whom MI had not yet been ruled out, in order to identify any acute coronary syndrome (including MI) on the index ED visit
2. They were derived prior to the advent of the newest high-sensitivity cardiac troponin assays
3. They fail to explicitly account for sex and pre-existing coronary disease, which are strong predictors of short-term adverse events

We will utilize data from two recently-completed Canadian ED chest pain prospective cohort studies to derive and validate novel risk scores to predict a patient’s risk of MACE after MI has been ruled out in the ED.

These cohorts, with a sample size of approximately 1100 patients each, will be used to derive and validate revised risk prediction tools that, we hypothesize, will have improved classification performance compared to existing tools, meaning that more patients will be able to be classified as low risk and not requiring additional testing, while maintaing excellent sensitivity for prediction of short term MACE.

The novel risk prediction tools that we will develop will innovate in the following ways:

1. We will include the right population, namely patients who have had MI ruled out in the ED, as this is the population representing a diagnostic challenge in terms of which patients need additional testing;
2. We will provide improved accounting for the difference in sex-specific MACE risk between men and women;
3. We will incorporate the latest generation cardiac troponin assay;
4. We will incorporate other commonly available biomarkers that have incremental prognostic utility in coronary artery disease;
5. We will undertake time-to-event analyses to suggest the optimal timing of additional cardiac testing;
6. We will provide granular, individualized risk estimates to guide decision-making around which patients need additional testing.

The knowledge product of this work will improve patient outcomes while also optimizing the appropriate utilization of objective cardiac testing modalities.