While new high-sensitivity cardiac troponin (hs-cTn) assays hold great promise for identifying acute myocardial infarction (AMI) early, physicians have struggled to interpret them. This issue of Strategies explores research that found a simple hs-cTn algorithm could rule-in or rule-out most patients who present to the emergency department with chest pain.

Adoption of hs-cTn assays in the U.S. has lagged behind Europe, in part due to questions about these assays’ positive predictive value that have held up regulatory approval from the U.S. Food and Drug Administration. Because these assays are so sensitive, they can pick up a variety of chronic or acute conditions other than AMI. At the same time, many physicians and laboratorians have urged the adoption of the new high-sensitivity assays, as patients with symptoms suggestive of AMI account for some 10% of emergency department (ED) visits. Recently, Swiss researchers proposed a strategy that could help clinicians make the most of hs-cTnT by employing an algorithm that successfully ruled-out or ruled-in 77% of patients in a multicenter study (Arch Intern Med 2012;172:1211–8).

The prospective multicenter study enrolled 872 patients with acute chest pain presenting to the ED. The cohort was part of the ongoing Advantageous Predictors of Acute Coronary Syndrome Evaluation (APACE) study. The researchers measured hs-cTnT in a blinded fashion at presentation at the ED and after 1 hour, and two independent cardiologists adjudicated patients’ final diagnosis. AMI was the final diagnosis in 17% of patients.

The researchers then developed an AMI rule-in/rule-out algorithm from a sample of 436 patients and validated it in the remaining 436 patients. The algorithm incorporated hs-cTnT baseline values and absolute changes within the first hour. Applying the algorithm to the validation cohort, 60% could be classified within 1 hour as rule-out, 17% as rule-in, and 23% as in an observational zone. Cumulative 30-day survival was 99.8% for patients classified as rule-out, 98.6% for patients in the observational zone, and 95.3% for those classified as rule-in.

The authors noted that their research resulted in four novel findings: first, that the proportion of patients with AMI continuously increases with hs-cTnT values; second, that the algorithm offered a “safe” rule-out and accurate rule-in for most patients; third, that this algorithm could significantly shorten the time for rule-out or rule-in; and fourth, that 30-day mortality was only 0.2% in patients ruled-out for AMI, “underscoring the suitability of these patients for early discharge.”

Their research points to the view that hs-cTnT should be interpreted as a quantitative variable rather than qualitative, and in this sense, the algorithm is a kind of compromise, said Christian Mueller, MD, an author of the study. “The best way to use any continuous biological information would be to use it as a continuous variable, but for clinical use, it makes a lot of sense to have broad categories and to translate that continuous information into two or three groups,” he said. “A more complex algorithm that added findings of an ECG or patient characteristics could also be incorporated into an electronic medical record with some simple software. This would be a very nice tool to get the most out of the quantitative information that you have from high-sensitivity troponin.” Mueller is a professor of medicine at University Hospital in Basel, Switzerland.

For rule-out of AMI, the optimal thresholds in the study achieved a 100% sensitivity and negative predictive value, and for rule-in, a specificity and positive predictive value of 97% and 84%, respectively. The rule-out criteria were a baseline hs-cTnT <12 ng/L and an absolute change within the first hour <3 ng/L. For rule-in of AMI, optimal thresholds were either a baseline hs-cTnT value at presentation ≥52 ng/L or an absolute change in hs-cTnT within the first hour ≥5 ng/L. Additional variables, such as age, sex, ischemic ECG changes, and time since onset of symptoms were not found to improve accuracy. Patients not fulfilling criteria for rule-in or for rule-out were classified as in the observational zone.

These findings will help U.S.-based clinicians apply the new generation of troponin assays and deal with concerns about hs-cTnT’s lower overall specificity for AMI, according to James Januzzi, MD, who was not associated with the study. “When you look at what the greatest advantage of the high-sensitivity troponins is relative to a conventional troponin, it’s the enhanced early sensitivity and consequently the assays’ early ability to exclude the diagnosis if in fact they are negative, while also recognizing the diagnosis at an earlier time, thus expediting triage as well,” Januzzi said. “While this algorithm is not a panacea for fixing the chest pain epidemic that we deal with on an everyday basis in the emergency department, it certainly could enhance care in a large percentage of cases.” Nearly a quarter of patients remain unclassified in the algorithm, and the study did not consider patients with unstable angina pectoris, he added. Januzzi is director of the cardiac intensive care unit at Massachusetts General Hospital in Boston and an associate professor of medicine at Harvard Medical School.

Januzzi also noted that the study lends support to the argument that absolute change is more informative than percent change for hs-cTnT. “There is some controversy in the lab world about this. Some people really advocate for a 50 percent rise or fall in troponin as being consistent with myocardial infarction,” he said. “The problem with this is that if you are down in the range of 14–49 ng/L—a category in which only 20 percent of patients had an MI—and go down by 50 percent, that may not be an acute MI, especially if other diagnoses responsible for these low level troponin values are present, such as heart failure. So this shows that the relative change is perfectly acceptable when you have big changes, but when you’re starting out at such a low concentration, a percent change may not be clinically meaningful.” He added that the Universal Definition of Myocardial Infarction global task force, of which he is a member, has yet to take a firm stance on percent change versus absolute change.

Before clinicians can start using an algorithm such as the authors propose, it should be externally validated in an independent, multicenter study, Mueller said. In fact, such a study is already underway, and it will prospectively validate the algorithm in 1,000 patients, he added.

Laboratorians may be familiar with recent problems Roche Diagnostics faced with cardiac troponin assays; however, these were not the same assays used in the study. Roche issued a Class I recall for lots 163176 and 163177 of its cardiac troponin I immunoassays in April of this year.

In an invited commentary, Kristin Newby, MD of Duke University Medical Center in Durham, N.C., called the study “an important step forward in application of hs-cTnT as a tool for triage of ED patients with possible MI” (Arch Intern Med. 2012;172:8–9). However, clinicians will likely need help from clinical decision support tools to make effective use of such an algorithm, she noted. “Finally, although touted as ‘simple’ by the authors, the need for multicomponent algorithms that are different for rule-in and rule-out and that vary by age group or other parameters will challenge application by busy clinicians unlikely to remember or accurately process the proposed algorithm. As such, it will be imperative that hs-cTnT algorithms, if validated, are built into clinical decision support layered onto electronic health records so that testing results are provided electronically to physicians along with the algorithmic interpretation to allow systematic application in triage and treatment.”