Objective: To determine the proportion of adverse events in patients discharged after ED assessment for possible acute coronary syndrome. Methods: Prospective observational cohort study enrolling consecutive patients presenting with symptoms suggestive of coronary syndrome. Main outcome was the proportion of adverse coronary events (defined a priori) within 30 days. Results: Of 2627 patients, 1819 (69%) were discharged without a diagnosis of coronary syndrome and 808 (31%) were admitted for further investigation and treatment. Of these, 385 (14.7%) were given a final diagnosis of acute coronary syndrome. On 30 day follow up, 18 of the discharged patients were diagnosed with acute coronary syndrome (0.7%; 95% confidence intervals [CI] 0.4–1.1%), 10 with unstable angina (0.4%; 95% CI 0.2–0.7%) and 8 with non-ST elevation myocardial infarction (0.3%; 95% CI 0.2–0.6%). There were no cases of ST elevation infarction or death. The sensitivity for diagnosis of acute coronary syndromes was 95.5% (95% CI 92.9–97.3%). Average length of stay was 7 h for discharged patients. Forty-six per cent of patients with diabetes and 47% with a past history of coronary disease were discharged. Subsequent outpatient stress testing was performed in 13.6%. Conclusions: In a large Australian ED, less than 1% of patients presenting with symptoms suggestive of coronary syndrome were discharged and subsequently had a 30 day adverse event. Reducing this proportion by admitting patients with traditional risk factors would markedly increase hospital workload. Opportunities exist to improve both the safety and efficiency of chest pain assessment in the ED.
Emergency Medicine Australasia Vol. 21, Issue 6, p. 455-464