A clinical trial of a chest-pain observation unit for patients with unstable angina. Chest Pain Evaluation in the Emergency Room (CHEER) Investigators

Abstract

Background: Nearly half of patients hospitalized with unstable angina eventually receive a non-cardiac-related diagnosis, yet 5 percent of patients with myocardial infarction are inappropriately discharged from the emergency department. We evaluated the safety, efficacy, and cost of admission to a chest-pain observation unit (CPU) located in the emergency department for such patients.

Methods: We performed a community-based, prospective, randomized trial of the safety, efficacy, and cost of admission to a CPU as compared with those of regular hospital admission for patients with unstable angina who were considered to be at intermediate risk for cardiovascular events in the short term. A total of 424 eligible patients were randomly assigned to routine hospital admission (a monitored bed under the care of the cardiology service) or admission to the CPU (where patients were cared for according to a strict protocol including aspirin, heparin, continuous ST-segment monitoring, determination of creatine kinase isoenzyme levels, six hours of observation, and a study of cardiac function). The CPU was managed by the emergency department staff. Patients whose test results were negative were discharged, and the others were hospitalized. Primary outcomes (nonfatal myocardial infarction, death, acute congestive heart failure, stroke, or out-of-hospital cardiac arrest) and use of resources were compared between the two groups.

Results: The 212 patients in the hospital-admission group had 15 primary events (13 myocardial infarctions and 2 cases of congestive heart failure), and the 212 patients in the CPU group had 7 events (5 myocardial infarctions, 1 death from cardiovascular causes, and 1 case of congestive heart failure). There was no significant difference in the rate of cardiac events between the two groups (odds ratio for the CPU group as compared with the hospital-admission group, 0.50; 95 percent confidence interval, 0.20 to 1.24). No primary events occurred among the 97 patients who were assigned to the CPU and discharged. Resource use during the first six months was greater among patients assigned to hospital admission than among those assigned to the CPU (P<0.01 by the rank-sum test).

Conclusions: A CPU located in the emergency department can be a safe, effective, and cost-saving means of ensuring that patients with unstable angina who are considered to be at intermediate risk of cardiovascular events receive appropriate care.