Implementation of a hs-cTnI Assay and Risk of Recurrent MI and Death in Patients with Suspected ACS

Editor's Note: This article is in response to Mills N, Churchhouse A, Lee K, et al. Implementation of a Sensitive Troponin I Assay and Risk of Recurrent Myocardial Infarction and Death in Patients With Suspected Acute Coronary Syndrome. JAMA 2011; 305:1210-1216.

The new high sensitivity Troponin I or T assays (hs-cTnI and hs-cTnT) identify more patients at risk for adverse cardiac events when compared to older assays. There is uncertainty, however, if the use of the hs-cTnI/hs-cTnT assays will lead to improved patient outcomes. This study by Mills et al. attempts to address this issue.

A total of 2,092 patients were evaluated in 2 phases in the Emergency Department at The Royal Infirmary in Edinburgh, Scotland. There were 1,038 patients studied in the validation phase, and 1,054 patients studied in the implementation phase in which a newer hs-cTnI assay (Abbott Architect) was used. In the implementation phase, values ≥ 0.05 ng/ml were recorded but only values ≥ 0.2 ng/ml were reported as abnormal to the clinicians. In the validation phase, values ≥ 0.05 ng/ml were reported to the clinicians as abnormal. All patient management decisions were left at the discretion of the responsible physicians.

In the entire group, hs-cTnI levels were <0.05 ng/ml in 1,340 (64%) patients, 0.05-0.019 ng/ml in 170 (8%) patients, and ≥ 0.2 ng/ml in 582 (28%) patients. Patients with hs-cTnI values between 0.05-0.019 ng/ml were treated differently in the two periods, when clinicians were aware of low-level hs-cTnI elevation. In the implementation phase those with hs-cTnI values 0.05-0.019 ng/ml were more likely to have a cardiology consultation (74% vs. 44%), undergo coronary angiography (46% vs. 20%), receive dual-antiplatelet therapy (58% vs. 27%), and receive statin therapy (80% vs. 58%) when compared to the validation phase, (p< 0.05 for all comparisons). The management of patients with hs-cTnI < 0.05 ng/ml and ≥ 0.2 ng/ml was unchanged following the reduction in the diagnostic threshold.

There was also a difference in outcomes in patients with hs-cTnI values between 0.05-0.19 ng/ml in the two periods. At 3 months there were 14 (16%) deaths in the validation phase compared to 4 (5%) deaths in the implementation phase (p<0.05), as well as more readmissions for acute myocardial infarction 18 (20%) compared to 5 (6%) (p<0.05).


The assessment of patients in the Emergency Department is clinically challenging. Many resources are used in the management of these patients and the vast majority (~ 85%) do not have acute coronary syndrome (ACS). Clinicians are keenly concerned about missing a patient with ACS, and older studies demonstrated that up to 2-5% of patients with ACS are not identified and inappropriately sent home, often with worse outcomes compared to admitted patients. The new hs-cTnI/hs-cTnT assays will clearly identify more patients with myocardial damage. The question is whether these patients are managed as if they have NSTEMI rather than a non-ACS elevation. The study by Mills et al., demonstrated that when clinicians were aware of intermediate hs-cTnI values (between 0.04-0.019 ng/ml) patients were managed differently, which was associated with lower mortality and less recurrent myocardial infarctions.

This provocative study does suggest that the use of a newer hs-cTnI assay may lead to identification of higher risk patients and improved patient outcomes. However, several aspects of this study need to be considered:

  1. This was a single-center historical control study that was not randomized, a problem with almost all studies evaluating biomarkers for treating possible ACS patients. In addition, there have been no multi-center randomized trials to show that the measurement of cTnI or cTnT leads to lower mortality or decreased MI when compared to CK-MB. Patients with low-level hs-cTnI elevations were treated more often with statins and clopidogrel, and underwent angiography more frequently in the implementation phase. Despite the increased use of coronary angiography, revascularization rates were not increased, nor were they treated with more beta-blockers, ACE inhibitors, or aspirin. Can all of the improvement in adverse cardiac events really be attributed to the use of more clopidogrel and statins?
  2. The identification of a very high-risk group with low-level hs-cTnI elevations is an important finding and consistent with other studies. In the validation phase patients with low-level hs-cTnI elevation had a 27% rate of death or recurrent myocardial infarction at 30 days. If the improved outcomes in this group, once the low-level hs-cTnI values were know by the clinician, was due to less missed ACS in the Emergency Department, further studies.will be needed to be elucidate this.
  3. The authors should be commended on a clever study design. The study suggests that the greater identification of high risk patients with a hs-cTnI assay will lead to improved outcomes. This single center study will need to be confirmed in other populations. A number of these patients with low-level hs-cTn elevation will not be diagnosed with ACS, but will receive further testing in their evaluation that they may not have received if they did not have an elevated cTn. This study took place in Scotland, where there are significant differences in the management of low risk chest pain patients; how these results might translate to the USA is uncertain.

Keywords: Scotland, Troponin I, Troponin T

< Back to Listings