Aims: Despite advances in the treatment of acute coronary syndromes based on randomized trial data and published guidelines, the extent to which such treatments are applied in practice remains uncertain. Data from clinical trials derive from selected geographical areas and in highly selected populations of patients, and hence may not reflect the overall population. The aim of the study was to investigate variations in hospital management and outcome using unselected data collected in the prospective Global Registry of Acute Coronary Events (GRACE).
Methods and results: The 95 hospitals in GRACE were organized into 18 population-based clusters in 14 countries. Information was recorded about patient management and outcome during hospitalization and after discharge. Data on treatments administered were analysed by baseline condition, hospital type, by the presence or absence of a catheterization laboratory, and by geographical region. Of 11543 patients, 44% had an admission diagnosis of unstable angina, 36% presented with myocardial infarction, 9% were admitted to rule out a myocardial infarction, 7% had chest pain and 4% were hospitalized for 'other cardiac' and 'non-cardiac' diagnoses. Of the total GRACE population 38% had a final diagnosis of unstable angina, 30% ST-segment elevation myocardial infarction, 25% non-ST-segment elevation myocardial infarction, and 7% of 'other cardiac' and 'non-cardiac' final diagnoses. The event rates for hospital death or reinfarction were six and 2% for non-ST-segment elevation myocardial infarction, seven and 3% for ST-segment elevation myocardial infarction, and 3% hospital death for unstable angina. The use of aspirin was similar across all hospital types and geographical regions. In contrast, the use of percutaneous coronary intervention and glycoprotein IIb/IIIa inhibitors was higher (P<0.0001) in teaching hospitals and hospitals with catheterization laboratories and was also higher in the United States. At discharge a higher percentage (P<0.0001) of patients received angiotensin-converting enzyme inhibitors in hospitals without catheterization laboratories. The use of statins was lower in non-teaching hospitals and in centres without a catheterization laboratory.
Conclusions: The GRACE study reveals substantial differences in the management of patients based on hospital type and geographical location. Further analyses will determine whether such variations translate into differences in longer term outcomes. GRACE provides a multinational reference for the implementation of therapies of proven efficacy.
Copyright 2002 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved.