We compare the ability of 5 established risk scores to identify patients with suspected acute coronary syndromes who are suitable for discharge after a modified single-presentation high-sensitivity troponin result. This was a prospective observational study conducted in a UK district general hospital emergency department. Consecutive adults recruited with suspected acute coronary syndrome for whom attending physicians determined evaluation with serial troponin testing was required. Index tests were definitions of low risk applied to modified Goldman, Thrombolysis in Myocardial Infarction (TIMI), Global Registry of Acute Cardiac Events (GRACE), History, ECG, Age, Risk Factors, Troponin (HEART), and Vancouver Chest Pain Rule risk scores, incorporating either high-sensitivity troponin T or I results. The endpoint was acute myocardial infarction within 30 days. A test sensitivity threshold for acute myocardial infarction of 98% was chosen. Clinical utility was defined as a negative predictive value greater than or equal to 99.5% and identification of greater than 30% suitable for discharge. Nine hundred fifty-nine patients underwent high-sensitivity troponin T analysis and 867 underwent high-sensitivity troponin I analysis. In the high-sensitivity troponin T group, 79 of 959 (8.2%) had an acute myocardial infarction and 66 of 867 (7.6%) in the high-sensitivity troponin I group. Two risk scores (GRACE <80 and HEART ≤3) did not have the potential to achieve a sensitivity of 98% with high-sensitivity troponin T, and 3 scores (Goldman ≤1, TIMI ≤1, and GRACE <80) with high-sensitivity troponin I. A TIMI score of 0 or less than or equal to 1 and modified Goldman score less than or equal to 1 with high-sensitivity troponin T, and TIMI score of 0 and HEART score of less than or equal to 3 with high-sensitivity troponin I had the potential to achieve a negative predictive value greater than or equal to 99.5% while identifying greater than 30% of patients as suitable for immediate discharge. With established risk scores, it may be possible to identify greater than 30% of patients suitable for discharge, with a negative predictive value greater than or equal to 99.5% for the diagnosis of acute myocardial infarction, using a single high-sensitivity troponin test result at presentation. There is variation in high-sensitivity troponin assays, which may have implications in introducing rapid rule-out protocols.
Read full abstract