Key result
Serial high-sensitivity cardiac troponin I testing using the COMPASS-MI algorithm achieved a maximum sensitivity of 98.6% for 30-day myocardial infarction or death, failing to meet the ≥99.0% benchmark required for safely ruling out patients.
Why the study?
Three major algorithms exist for serial hsTnI testing, but comparative diagnostic estimates and whether one is superior across different ED populations remained unknown.
Does serial high-sensitivity cardiac troponin I testing using ESC, High-STEACS, or COMPASS-MI algorithms improve diagnostic accuracy for 30-day MI or death in ED patients?
Population
2966 undifferentiated ED patients (Cohort-1) and 935 ED patients with suspected ACS (Cohort-2)
Comparison
ESC vs High-STEACS vs COMPASS-MI serial hsTnI algorithms
Design
Blinded prospective cohort study
Follow-up
30 days
Authors
Loading...
None of the three major hsTnI algorithms (ESC, High-STEACS, COMPASS-MI) achieved the proposed sensitivity benchmark of ≥99% for 30-day MI or death in ED patients.
Observational (n=3,901)
Physicians blinded to hsTnI results
Yes
Does serial high-sensitivity cardiac troponin I testing using ESC, High-STEACS, or COMPASS-MI algorithms improve diagnostic accuracy for 30-day MI or death in ED patients?
Effect estimate: Sensitivity 98.6% (95% CI 95.1-99.8)
None of the three major hsTnI algorithms (ESC, High-STEACS, COMPASS-MI) achieved the proposed sensitivity benchmark of ≥99% for 30-day MI or death in ED patients.
Kavsak et al. (2021) conducted an observational in Suspected Acute Coronary Syndrome (n=3,901). Serial high-sensitivity cardiac troponin I (hsTnI) algorithms (COMPASS-MI, ESC 0/2h, High-STEACS) vs. Diagnostic benchmarks (sensitivity ≥ 99.0%, specificity ≥ 90.0%) was evaluated on 30-day myocardial infarction or all-cause death (Sensitivity of COMPASS-MI algorithm in Cohort-2) (Sensitivity 98.6%, 95% CI 95.1-99.8). Serial high-sensitivity cardiac troponin I testing using the COMPASS-MI algorithm achieved a maximum sensitivity of 98.6% for 30-day myocardial infarction or death, failing to meet the ≥99.0% benchmark required for safely ruling out patients.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: