Chest pain after recent stress test: Is there a warranty?

Abstract

Study objectives: Emergency physicians often incorporate recent stress test results into their decisionmaking process for the evaluation and disposition of chest pain patients, which is based on studies of prognosis after stress testing. However, the subset of patients who present to an emergency department (ED) with chest pain and a history of a normal stress test is not adequately addressed in the literature. We seek to determine whether a history of a recent normal stress test result in patients presenting with chest pain to the ED is associated with a low risk (<1%) of major acute cardiac events. Methods: Patients presenting to our ED with a chief complaint of chest discomfort between January 1, 1998, and September 30, 2002, and who had undergone a stress test at our institution within 2 years were included. The primary endpoint was acute myocardial infarction on ED visit or death within 30 days. The diagnosis of acute myocardial infarction was based on cardiac marker elevation or ECG. All ECGs were reread for the purpose of this study by an investigator blinded to all other clinical data. Previous stress tests included treadmill studies without imaging, treadmill studies with imaging, and pharmacologic imaging studies. Treadmill study results were considered normal if there were no ST-segment changes and no abnormalities of functional capacity or heart rate dynamics during or after exercise. Imaging study results were considered normal if there was no evidence of myocardial ischemia or scar. Results: There were 1,857 eligible patients, of whom 283 had treadmill tests without imaging (79 or 27% were normal), 750 had treadmill tests with imaging (235 or 31% were normal), and 824 had pharmacologic tests (448 or 54% were normal). There were 190 patients (10%) who had the primary endpoint; only 2 patients died within 30 days. The history of a normal stress test was associated with a lower risk of the primary endpoint (50/762 or 7% versus 140/1095 or 13%, odds ratio 0.48, 95% confidence interval [CI] 0.34 to 0.67, <i>P</i><.0001). Among 449 patients who had a normal stress test result and on initial presentation had a nondiagnostic ECG and normal biomarkers, only 9 (2.0%, 95% CI 1.1% to 3.8%) had the primary endpoint. Among 508 patients who, on initial presentation, had a nondiagnostic ECG and normal biomarkers but had a history of an abnormal stress test result, 16 (3.2%, 95% CI 2.0% to 5.1%) had the primary endpoint. Conclusion: Among patients presenting with chest pain to the ED and a history of stress testing within the past 2 years, a normal stress test result was associated with a markedly reduced risk of having a major cardiac endpoint. However, with the possible exception of patients presenting with initially normal biomarkers and ECG findings, the absolute risk of 7% was not reassuringly low enough to justify early discharge.