High-Sensitivity Troponin for Rapid MI Rule-In/Rule-Out

Dirk Westermann, MD, PhD


September 21, 2015

This feature requires the newest version of Flash. You can download it here.

Hi. Good day from London, at the European Society of Cardiology (ESC) meeting. My name is Dirk Westermann. I'm a cardiologist from Hamburg University and I just presented the BACC trial.[1] BACC is the acronym for Biomarkers in Acute Cardiovascular Care.

We all know that the number of patients presenting in the emergency department with acute chest pain suggestive of myocardial infarction is rising. We see many patients daily. Therefore, we need to safely and rapidly rule out, but also rule in, myocardial infarction in order to initiate evidence-based treatment for those with acute myocardial infarction and to also discharge those without acute cardiac conditions. Therefore, safe and rapid rule-out is important. This may also limit the scarce medical resources at emergency departments.

The recent ESC guidelines[2] ask to measure high-sensitivity troponin directly at admission and with a delay of 3 hours, and we should take the 99th percentile as the cut-off for elevated values. Today they presented new NSTEMI [non-ST-elevation myocardial infarction] guidelines.[3] The ESC changed them and now they allow faster algorithms with lower cutoffs.

In BACC, we aimed to investigate the faster rule-in and rule-out at 1 hour compared with the 3 hours approach, and the low and more sensitive troponin I cutoff compared with the 99th percentile. We identified 1045 patients presenting with acute chest pain suggestive of acute myocardial infarction at our emergency rooms. Clinically, we detected NSTEMI by the ESC guidelines. We measured troponin T directly at admission and after 3 hours. Integrating clinical judgment, imaging, and ECG, we made the final diagnosis of NSTEMI.

For study reasons, we measured troponin I, which is another high-sensitivity troponin assay, directly at admission and after 1 hour. We did not wait for 3 hours but measured it after 1 hour. We then calculated the best-performing cutoff value in BACC, which was 6 ng/L. It had the highest negative predictive value, the lowest number of false-negative patients, and, importantly, it was above 5.2 ng/L, which is the 10% coefficient of variation for this specific assay, an important marker of assay precision. We used this cutoff of 6 ng/L and applied it to our cohort to rule out myocardial infarction. By doing so, we created a rule-out algorithm that was troponin I < 6 ng/L directly at admission and < 6 ng/L after 1 hour. We identified 402 of the 1045 patients who could be discharged directly after 1 hour.

The negative predictive value was very high after 1 hour: 99.7% for NSTEMI type 1. Importantly, it was as safe as the 3-hour approach and there was no loss of clinical safety. When we compared our new, more sensitive cutoff with the 99th percentile, we found that the 99th percentile performed less well and with lower negative predictive values. When we looked for follow-up mortality, it was very low, with only three fatal events when we chose the cutoff of 6 ng/L and used the 1-hour algorithm. If we would have discharged patients based on the 99th percentile, we would have seen 12 fatal events. Therefore, our new algorithm is not only faster, but it's also safer. We validated our algorithm in two independent cohorts of patients with acute chest pain, one from Europe and one from Australia. Both cohorts validated the safety of that algorithm.

Finally, we applied those cutoff values to the general population. We used the BiomarCaRE cohort, a harmonized cohort incorporating several European cohorts from the general population. We measured troponin I in roughly 75,000 individuals from the general population. Follow-up mortality was lower in those with values < 6 ng/L compared with those who had values below < 27 ng/L, which is the recently suggested 99th percentile.

Therefore, we can conclude that ruling out myocardial infarction can be achieved after 1 hour. The lower, more sensitive cutoff performs better with lower follow-up mortality. Slightly elevated troponin I levels imply cardiovascular risk. This troponin I assay is not approved by the US Food and Drug Administration yet. We'll see whether that changes, but I think it's an important part of better care for our patients with acute chest pain.

Thank you very much for your attention.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: