Log in to MyACC
Performance of ESC 0/1-, 0/2-, and 0/3-Hour Algorithms for AMI
Quick Takes
- This is a meta-analysis comparing the performance of the 0/1-, 0/2-, and 0/3-hour high-sensitivity cardiac troponin algorithms in ruling out AMI in 32 studies and 30,066 patients presenting to the ED with chest pain.
- The ESC 0/1- and 0/2-hour algorithms have higher sensitivities and negative predictive values than the 0/3-hour algorithm for AMI.
- Given a pretest probability of 12% (median prevalence of AMI across the studies), 120 of 1,000 tested patients would receive a final diagnosis of AMI; of these 120 patients, we can be confident that no more than 3 may test false negative according to the 0/1- and 0/2-hour algorithms and up to 15 may test false negative according to the 0/3-hour algorithm. Of the 880 without AMI, 18-115 patients may test false positive.
Study Questions:
How do the 0/1-, 0/2-, and 0/3-hour high-sensitivity cardiac troponin (hs-cTn) algorithms compare in ruling out acute myocardial infarctions (AMIs) in patients presenting to the emergency department (ED) with chest pain?
Methods:
This study is a meta-analysis comparing the diagnostic accuracies of the European Society of Cardiology (ESC) 0/1-, 0/2-, and 0/3-hour hs-cTn based algorithms. A total of 32 studies (20 cohorts) with 30,066 patients were analyzed. Primary data were obtained from investigators of 16 cohorts and aggregate data were extracted from four cohorts. The primary outcome was diagnostic accuracy (sensitivity, specificity, negative predictive value [NPV], positive predictive value [PPV]) with index admission acute MI (AMI) (type 1 or type 2 MI). The secondary outcomes were 30-day mortality and proportion of patients triaged toward rule-out, observation, or rule-in. Sources of heterogenicity considered included geographic location, prevalence of AMI, proportion of patients presenting with chest pain, study design, sample size, and method of adjudication of AMI.
Results:
All studies prospectively recruited patients presenting to the ED or chest pain unit with symptoms suggestive of AMI. Most studies had low risk of selection bias because they enrolled patients consecutively. The 0/1-hour algorithm had a pooled sensitivity of 99.1% (95% confidence interval [CI], 98.5-99.5%) and NPV of 99.8% (CI, 99.6-99.9%) for ruling out AMI. The 0/2-hour algorithm had a pooled sensitivity of 98.6% (CI, 97.2-99.3%) and NPV of 99.6% (CI, 99.4-99.8%). The 0/3-hour algorithm had a pooled sensitivity of 93.7% (CI, 87.4-97.0%) and NPV of 98.7% (CI, 97.7-99.3%). All three algorithms had similar specificities and PPVs for ruling in AMI. Diagnostic performance was similar across the hs-cTnT (Elecsys; Roche), hs-cTnI (Architect; Abbott), and hs-cTnI (Centaur/Atellica; Siemens) assays.
Conclusions:
The ESC 0/1- and 0/2-hour algorithms have higher sensitivities and NPVs than the 0/3-hour algorithm for AMI in patients presenting to the ED with chest pain.
Perspective:
With the advent of hs-cTnT testing, numerous ED triage algorithms have been tested to improve the sensitivity and timing of AMI diagnosis in the ED. Accordingly, the ESC recommendations have rapidly evolved, now recommending the 0/1- and 0/2-hour algorithms as first and second choices and the 0/3-hour algorithm as an alternative based on logistical feasibility. This meta-analysis comparing all three algorithms shows that the 0/1- and 0/2-hour algorithms had consistently high sensitivities and NPV across different population demographic characteristics. The NPVs for all studies, regardless of AMI prevalence, were close to or above 99%, which approximates accepted safety estimates of 1% miss rate. By contrast, the 0/3-hour algorithm had a lower sensitivity and NPV than the 0/1- and 0/2-hour algorithms. This difference in performance despite the added triage time for the 0/3-hour algorithm is likely due to the lower rule-out threshold and delta for cTn levels to rule out AMI. Another reason for the differences is that studies not implementing clinical criteria such as the GRACE score had a lower sensitivity for the ESC 0/3-hour algorithm, while studies incorporating clinical criteria with the 0/3-hour algorithm had similar sensitivities to the 0/1- and 0/2-hour algorithms. Overall, this study supports the use of 0/1- or 0/2-hour hs-cTn algorithms in the triage of patients with chest pain. The ESC 0/3-hour algorithms should only be used in conjunction with clinical criteria.
Clinical Topics: Acute Coronary Syndromes, Cardiovascular Care Team, Prevention, ACS and Cardiac Biomarkers
Keywords: Acute Coronary Syndrome, Algorithms, Biomarkers, Cardiology Interventions, Chest Pain, Emergency Service, Hospital, Myocardial Infarction, Secondary Prevention, Triage, Troponin T
< Back to Listings
