Direct Comparison of 4 Very Early Rule-Out Strategies for Acute Myocardial Infarction Using High-Sensitivity Cardiac Troponin I

Background: Four strategies for very early rule-out of acute myocardial infarction using high-sensitivity cardiac troponin I (hs-cTnI) have been identified. It remains unclear which strategy is most attractive for clinical application. Methods: We prospectively enrolled unselected patients presenting to the emergency department with symptoms suggestive of acute myocardial infarction. The final diagnosis was adjudicated by 2 independent cardiologists. Hs-cTnI levels were measured at presentation and after 1 hour in a blinded fashion. We directly compared all 4 hs-cTnI–based rule-out strategies: limit of detection (LOD, hs-cTnI<2 ng/L), single cutoff (hs-cTnI<5 ng/L), 1-hour algorithm (hs-cTnI<5 ng/L and 1-hour change<2 ng/L), and the 0/1-hour algorithm recommended in the European Society of Cardiology guideline combining LOD and 1-hour algorithm. Results: Among 2828 enrolled patients, acute myocardial infarction was the final diagnosis in 451 (16%) patients. The LOD approach ruled out 453 patients (16%) with a sensitivity of 100% (95% confidence interval [CI], 99.2%–100%), the single cutoff 1516 patients (54%) with a sensitivity of 97.1% (95% CI, 95.1%–98.3%), the 1-hour algorithm 1459 patients (52%) with a sensitivity of 98.4% (95% CI, 96.8%–99.2%), and the 0/1-hour algorithm 1463 patients (52%) with a sensitivity of 98.4% (95% CI, 96.8%–99.2%). Predefined subgroup analysis in early presenters (≤2 hours) revealed significantly lower sensitivity (94.2%, interaction P =0.03) of the single cutoff, but not the other strategies. Two-year survival was 100% with LOD and 98.1% with the other strategies ( P <0.01 for LOD versus each of the other strategies). Conclusions: All 4 rule-out strategies balance effectiveness and safety equally well. The single cutoff should not be applied in early presenters, whereas the 3 other strategies seem to perform well in this challenging subgroup. Clinical Trial Registration: URL: http://www.clinicaltrials.gov . Unique identifier: NCT00470587.