@article {Roose001682, author = {Andreas Roos and Martin J Holzmann}, title = {Use of historical high-sensitivity cardiac troponin T levels to rule out myocardial infarction}, volume = {8}, number = {1}, elocation-id = {e001682}, year = {2021}, doi = {10.1136/openhrt-2021-001682}, publisher = {Archives of Disease in childhood}, abstract = {Objective Several high-sensitivity cardiac troponin (hs-cTn)-based strategies exist for rule-out of myocardial infarction (MI). It is unknown whether historical hs-cTnT concentrations can be used. This study aim to evaluate the performance of a rule-out strategy based on the European Society of Cardiology (ESC) 0/1-hour algorithm, using historical hs-cTnT concentrations.Methods All visits among patients with chest pain in the emergency department at nine different hospitals in Sweden from 2012 to 2016 were eligible (221 490 visits). We enrolled patients with a 0-hour hs-cTnT of \<12 ng/L, a second hs-cTnT measured within 3.5 hours, and >=1 historical hs-cTnT available. We calculated the risks of MI and all-cause mortality using two rule-out strategies: (1) a delta hs-cTnT of \<3 ng/L between the 0-hour hs-cTnT and the second hs-cTnT (modified ESC algorithm) and (2) a historical hs-cTnT \<12 ng/L and a delta hs-cTnT of \<3 ng/L in relation to the 0-hour hs-cTnT (historical-hs-cTnT algorithm).Results A total of 8432 patients were included, of whom 84 (1.0\%) had an MI. The modified ESC algorithm triaged 8100 (96\%) patients toward ruled-out, for whom 30-day MI risk and negative predictive value (NPV) for MI (95\% CI) were 0.4\% (0.3\% to 0.6\%) and 99.6\% (99.4\% to 99.7\%), respectively. The historical-hs-cTnT algorithm ruled out 6700 (80\%) patients, with a 30-day MI risk of 0.5\% (0.4\% to 0.8\%) and NPV of 99.5\% (99.2\% to 99.6\%).Conclusions The application of algorithm resulted in similar MI risk and NPV to an established algorithm. The usefulness of historical hs-cTnT concentrations should merit further attention.Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information. The dataset was anonymised, so that no unique patient could be identified. No permitted commercial reuse of data.}, URL = {https://openheart.bmj.com/content/8/1/e001682}, eprint = {https://openheart.bmj.com/content/8/1/e001682.full.pdf}, journal = {Open Heart} }