Purpose: Addressing the increasingly recognized, yet unmet clinical need for rapid rule-out of acute myocardial infarction (AMI), several novel strategies have been developed. Due to the lack of direct comparisons in the same dataset, selection of the best strategy for clinical practice is challenging. We therefore aimed to directly compare the safety and efficacy of two previously defined strategies: LOD (Undetectable levels of high-sensitivity cardiac troponin (hs-cTn) T at presentation) in combination with hs-cTnT 1h-algorithm versus 2h-ADP rule-out strategy (0h and 2h levels≤99th percentile, no significant ECG changes and TIMI Score of ≤1).

Methods: In a prospective international multicentre diagnostic study enrolling 1697 patients presenting with suspected AMI to the emergency department, the final diagnosis of AMI was adjudicated by two independent cardiologists using all available clinical information including serial hs-cTnT concentrations. Safety was quantified as the negative predictive value (NPV) for AMI in the rule-out zone of the respective rule-out strategies. Efficacy was quantified as the percentage of the overall cohort assigned to the rule-out zone by the respective strategy. Both strategies were applied using the two best-validated hs-cTn assays (hs-cTnT Roche: LOD <5ng/l; 1h algorithm: 0h<12ng/l and Δ0–1h<3ng/l and 99th percentile =14ng/l; hs-cTnI Abbott: LOD<2ng/l; 1h-algorithm 0h <5ng/L and Δ0–1h<2ng/l; 99th percentile=26.2ng/l) to ensure that findings are independent from the hs-cTn assay used. As both strategies should only be applied once ST-elevation MI (STEMI) has been excluded by the initial ECG, STEMI patients were excluded from the analysis.

Results: Acute myocardial infarction was the final diagnosis in 16% of patients. The safety was very high and comparable with both algorithms (LOD+1h-Algorithm NPV 99.9%, 95% CI 99.5–100% versus 2h ADP: NPV 100%, 95% CI 99.4–100% p=ns). Regarding efficacy, 1h-algorithm +LOD allowed rule-out in 60%of patients versus 38% with 2h ADP (p<0.001). Using hs-cTnI, the safety was very high and comparable with both algorithms (LOD+1h-Algorithm NPV 99.4%, 95% CI 98.6–99.9% versus 2h ADP: NPV 100%, 95% CI 99.3–100% p=ns). Regarding efficacy 1h-algorithm +LOD allowed rule-out in 53%of patients versus 38% with 2h ADP (p<0.001).

Conclusion: Both investigated rule-out strategies allow a safe rule-out of AMI. However the combination of LOD+1h-algorithm has a significantly higher efficacy allowing the rule out of more than double of patients and has the obvious advantage of allowing rule-out already after 1h and without the need of TIMI Score calculation.