Application of High-Sensitivity Troponin in Suspected Myocardial Infarction

Johannes T. Neumann, Raphael Twerenbold, Francisco Ojeda, Nils A. Sörensen, Andrew R. Chapman, Anoop S. V. Shah, Atul Anand, Jasper Boeddinghaus, Thomas Nestelberger, Patrick Badertscher, Arash Mokhtari, John W. Pickering, Richard W. Troughton, Jaimi Greenslade, William Parsonage, Matthias Mueller-Hennessen, Tommaso Gori, Tomas Jernberg, Niall Morris, Christoph LiebetrauChristian Hamm, Hugo A. Katus, Thomas Münzel, Ulf Landmesser, Veikko Salomaa, Licia Iacoviello, Marco M. Ferrario, Simona Giampaoli, Frank Kee, Barbara Thorand, Annette Peters, Rossana Borchini, Torben Jørgensen, Stefan Söderberg, Susana Sans, Hugh Tunstall-Pedoe, Kari Kuulasmaa, Thomas Renné, Karl J. Lackner, Andrew Worster, Richard Body, Ulf Ekelund, Peter A. Kavsak, Till Keller, Bertil Lindahl, Philipp Wild, Evangelos Giannitsis, Martin Than, Louise A. Cullen, Nicholas L. Mills, Christian Mueller, Tanja Zeller, Dirk Westermann, Stefan Blankenberg (Lead / Corresponding author)

Research output: Contribution to journalArticlepeer-review

229 Citations (Scopus)


Background: Data regarding high-sensitivity troponin concentrations in patients presenting to the emergency department with symptoms suggestive of myocardial infarction may be useful in determining the probability of myocardial infarction and subsequent 30-day outcomes.

Methods: In 15 international cohorts of patients presenting to the emergency department with symptoms suggestive of myocardial infarction, we determined the concentrations of high-sensitivity troponin I or high-sensitivity troponin T at presentation and after early or late serial sampling. The diagnostic and prognostic performance of multiple high-sensitivity troponin cutoff combinations was assessed with the use of a derivation–validation design. A risk-assessment tool that was based on these data was developed to estimate the risk of index myocardial infarction and of subsequent myocardial infarction or death at 30 days.

Results: Among 22,651 patients (9604 in the derivation data set and 13,047 in the validation data set), the prevalence of myocardial infarction was 15.3%. Lower high-sensitivity troponin concentrations at presentation and smaller absolute changes during serial sampling were associated with a lower likelihood of myocardial infarction and a lower short-term risk of cardiovascular events. For example, high-sensitivity troponin I concentrations of less than 6 ng per liter and an absolute change of less than 4 ng per liter after 45 to 120 minutes (early serial sampling) resulted in a negative predictive value of 99.5% for myocardial infarction, with an associated 30-day risk of subsequent myocardial infarction or death of 0.2%; a total of 56.5% of the patients would be classified as being at low risk. These findings were confirmed in an external validation data set.

Conclusions: A risk-assessment tool, which we developed to integrate the high-sensitivity troponin I or troponin T concentration at emergency department presentation, its dynamic change during serial sampling, and the time between the obtaining of samples, was used to estimate the probability of myocardial infarction on emergency department presentation and 30-day outcomes.
Original languageEnglish
Pages (from-to)2529-2540
Number of pages12
JournalNew England Journal of Medicine
Issue number26
Publication statusPublished - 27 Jun 2019


  • Adult
  • Aged
  • Biomarkers/blood
  • Cohort Studies
  • Emergency Service, Hospital
  • Female
  • Humans
  • Male
  • Middle Aged
  • Myocardial Infarction/blood
  • Prognosis
  • Risk Assessment/methods
  • Sensitivity and Specificity
  • Troponin I/blood
  • Troponin/blood


Dive into the research topics of 'Application of High-Sensitivity Troponin in Suspected Myocardial Infarction'. Together they form a unique fingerprint.

Cite this