Identification of very low risk chest pain using clinical data in the emergency department

Background: Evaluation of chest pain of uncertain origin in the emergency department is a challenge. Chest pain units, involving non-invasive stress testing, have logistic constraints. Our aim was to identify very low risk patients for early discharge using clinical data.

Methods: A total of 772 patients were studied. Ischemia in the electrocardiogram, troponin elevation or history of ischemic heart disease, were exclusion criteria. The primary end point was 30 day cardiac events (death, myocardial infarction or revascularization). The secondary end point was 1 year major events (death or myocardial infarction).

Results: The primary and secondary end point rates were 123 (18%) and 31 (4%). Predictive variables for the primary end point were typical chest pain (OR=1.8, p=0.007), ≥ 2 pain episodes in last 24h (OR=3.4, p=0.0001), age ≥ 55 years (OR=1.8, p=0.03), male (OR=2.2, p=0.001), diabetes (OR=1.8, p=0.01) and family history of ischemic heart disease (OR=2.0, p=0.02). A very low risk category could be distinguished (<2 predictors, n=114) that showed only 3 (2.6%) events at 30 days (all 3 revascularizations), compared with 120 (18%) in the remaining patients (p=0.0001). The very low risk criteria had 97% negative predictive for 30 day cardiac events. No very low risk patient presented major events at 1 year compared with 31 (4.7%) in the remaining patients (p=0.009).

Conclusion: In patients presenting to the emergency department with chest pain of uncertain origin and without prior ischemic heart disease, very low risk patients can be identified using clinical data. These patients could be quickly discharged without further non-invasive stress testing.