Can Potential Acute MI Patients be Triaged Faster?

Abstract & Commentary

By Michael H. Crawford, MD, Professor of Medicine, Chief of Clinical Cardiology, University of California, San Francisco. Dr. Crawford reports no financial relationships relevant to this field of study.

This article originally appeared in the February 2012 issue of Clinical Cardiology Alert. It was peer reviewed by Ethan Weiss, MD. Dr. Weiss is Assistant Professor of Medicine, Division of Cardiology and CVRI, University of California, San Francisco. Dr. Weiss is a scientific advisory board member for Bionovo.

Source: Keller T, et al. Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infraction. JAMA 2011;306:2684-2693.

Newer, more sensitive troponin assays have the potential to identify acute myocardial infarction (AMI) earlier, but some detect troponin in 50% of normal populations, which renders them clinically useless. This study evaluates the use of troponin kinetics clinically to separate AMI patients from those with chronically elevated troponin levels. They enrolled 1818 consecutive patients suspected of having AMI and measured high sensitivity troponin I (hsTnI) and conventional troponin I (cTnI) at admission and 3 and 6 hours after. Patients were followed for 30 days and a final diagnosis of AMI was made based on current guidelines by blinded cardiologists. Critical to the diagnosis was at least a 20% rise and fall in a conventional troponin assay (I or T). The final diagnosis of MI was made in 413 (23%), of whom 56 (14%) had ST elevation MI. AUC values for the ROC curves were highest for hsTnI (0.96) followed by cTnI (0.92) on admission. The hsTnI at admission had a sensitivity of 82% and a negative-predictive value (NPV) of 95%, whereas for cTnI the values were 79% and 94%. The positive-predictive value (PPV) on admission for hsTnI was 75% and for cTnI was 81%. Samples at 3 hours improved the sensitivity of hsTnI to 98% with a NPV of 99% and identical values were observed for cTnI. Using the criteria of a hsTnI > 99th percentile at admission and a change > 266% at 3 hours improved PPV for both assays to 96%. The authors concluded that TnI measurements on admission and 3 hours later may facilitate the early triage of suspected AMI patients.


This is a relatively large observational study that compares hsTnI to cTnI and compares admission values to 3 hours post admission values and the change from admission to 3 hours for diagnosing AMI. The data analysis is comprehensive and quite complicated. Also, other biomarkers were evaluated in the study, but none, alone or in combination, were better than troponin. The study focused on suspected AMI patients because of chest pain symptoms. Only 14% of the MIs proved to be STEMIs, so the study mainly was about identifying non-STEMIs. Several conclusions can be made. First, at the time of admission hsTnI and cTnI were similar for ruling out AMI (NPV 95% vs. 94%). For diagnosing AMI at admission cTnI was a little better than hsTnI (PPV 81 vs. 75%). So for initial triage there is no real value of hsTnI over cTnI. Second, most low-risk patients will be held until a second troponin is done to improve diagnostic accuracy. Three hours hsTnI had a NPV of 99%, as did cTnI. So again there was no added benefit of hsTnI. Third, an evaluation of the change in troponin from admission to 3 hours showed that the PPV for both assays was increased to 96%, again demonstrating no advantage to using hsTnI. Fourth, the only subgroups that showed an advantage for hsTnI were patients with initially normal troponin and patients known to have presented very early after symptom onset (< 2 hrs). Fifth, using hsTnI routinely will result in a higher false-positive rate at admission as compared to cTnI. In summary, at this point I do not see any reason to switch to hsTnI.

There are some limitations to this study. Since troponin values were used to adjudicate the AMI diagnosis, there is the risk of incorporation bias, which would tend to increase the accuracy of troponin tests. This population was unique in at least two regards. There was a relatively high rate of AMI (23%) compared to other studies. This would also tend to increase accuracy. Also, almost all of the patients were white Europeans; so the results may not be applicable to other ethnic groups. Overall, this is a robust study which will undoubtedly influence the application of hsTnI assays in clinical settings.