Effect of Implementation of HEART Chest Pain Protocol on Emergency Department Disposition, Testing and Cost

Symptoms concerning for acute coronary syndromes (ACS) such as chest pain and dyspnea are some of the most common reasons for presenting to an emergency department (ED). The HEART score (history, electrocardiogram, age, risk factors and troponin) was developed and has been externally validated in an emergency setting to determine which patients with chest pain are at increased risk for poor outcomes. Our hospital adopted a HEART score-based protocol in late 2015 to facilitate the management and disposition of these patients. In this study we aimed to analyze the effects of the adoption of this protocol. Prior studies have included only patients with chest pain. We included both patients with chest pain and patients with only atypical symptoms.

This was a retrospective chart review of two cohorts. We identified ED charts from six-month periods prior to and after adoption of our HEART score-based protocol. Patients in whom an electrocardiogram and troponin were ordered were eligible for inclusion. We analyzed data for patients with typical symptoms (chest pain) and atypical symptoms both together and separately.

We identified 1546 charts in the pre-adoption cohort and 1623 in the post-adoption cohort that met criteria. We analyzed the first 900 charts in each group. Discharges from the ED increased (odds ratio [OR[1.56, P<.001), and admissions for cardiac workup decreased (OR 0.46, P <.001). ED length of stay was 17 minutes shorter (P = .01). Stress testing decreased (OR 0.47, P<.001). We estimate a cost savings for our hospital system of over $4.5 million annually. There was no significant difference in inpatient length of stay or catheterization rate. When analyzing typical and atypical patients separately, these results held true.

After adoption of a HEART score-based protocol, discharges from the ED increased with a corresponding decrease in admissions for cardiac evaluations as well as cost. These effects were similar in patients presenting without chest pain but with presentations concerning for ACS.