ESC Algorithm Safely Rules Out NSTEMI, but Caution May Be Warranted in the US

A European Society of Cardiology (ESC)-recommended 0/1-hour algorithm using high-sensitivity cardiac troponin measurements safely and accurately triages patients with suspected NSTEMI in the emergency department (ED), according to a large validation study conducted in Europe. US physicians, however, are wary of applying the results to Americans with chest pain.

Measuring troponins at presentation and 1 hour later safely ruled out NSTEMI, with a negative predictive value of 99.7% to 99.8%, and effectively triaged 67% to 75% of patients with chest pain to either the rule-in or rule-out group, researchers led by Raphael Twerenbold, MD (University Hospital Basel, Switzerland), and Johannes Tobias Neumann, MD (University Hospital Hamburg-Eppendorf, Hamburg, Germany), report.

Twerenbold initially reported the study results at last year’s ESC Congress in Barcelona, Spain; they were published online this week ahead of the August 7, 2018, issue of the Journal of the American College of Cardiology.

“It’s a frequent scenario that patients present with symptoms suggestive of myocardial infarction,” Twerenbold told TCTMD. Applying this algorithm “substantially speeds up and facilitates the management of these patients, which is good for the patients because the time they have to wait with the uncertainty and the anxiety until they get exclusion of myocardial infarction or the assignment to high risk is reduced.”

The approach also allows for earlier initiation of treatment in patients who need it and relieves overcrowding in the ED by more rapidly determining whether patients are eligible to be sent home or need to be admitted, he added.

Publication of the results, however, has stirred up some consternation on the other side of the Atlantic. US physicians should be cautious about applying these results to their patients with chest pain, according to W. Frank Peacock, MD (Baylor College of Medicine, Houston, TX).

“If emergency physicians in the United States start doing it, they’ll kill people,” he told TCTMD.

He raised issues around differences between the United States and Europe in terms of the types of patients presenting with chest pain, high-sensitivity assay availability (the first high-sensitivity troponin assay was approved in the US in January 2017), and the cut points used to triage patients. In addition, he said, there was no information provided about the care patients received after they left the ED.

Underscoring the dangers of extrapolating findings like these to different populations, Peacock pointed to a prior study that found unacceptable performance of this same algorithm when used in centers in New Zealand, Australia, and Canada.

The patient population is key when interpreting how these results might apply in a US setting, Peacock said, noting that very few hospitals in the United States have a rate of MI among all chest pain patients similar to what was seen in this study (17% diagnosed with NSTEM). In addition, he pointed out, high-sensitivity troponin assays are not yet widely available.

“This is really good information for people in Europe and for people who study troponin in the United States,” Peacock continued. “It is not good for the people who practice emergency medicine in the United States until you can put all the pieces together. Do not misjudge this or you do so at your patients’ peril.”

Good Performance Even in Early Presenters

The emergence of high-sensitivity cardiac troponin assays has allowed emergency physicians to more rapidly triage patients with symptoms suggestive of acute MI. The 0/1-hour algorithm is recommended by the ESC as an option in its guidelines on management of ACS in patients without persistent ST-segment elevation, but some concerns have been raised about its use, particularly in patients presenting within 3 hours of chest pain onset (early presenters).

For their analysis, Twerenbold, Neumann, and colleagues evaluated data from two studies—APACE and BACC—on patients with symptoms suggestive of NSTEMI who were prospectively enrolled at EDs in six European countries; 30% were early presenters. Overall, 17% of patients were ultimately diagnosed with NSTEMI.

Two high-sensitivity troponin assays were used: the Elecsys troponin T assay (Roche Diagnostics) and the Architect troponin I assay (Abbott).

The numbers in the published paper closely match those presented last year. Among the 4,368 patients with serial troponin T measurements available, the 0/1-hour approach safely ruled out NSTEMI with a negative predictive value of 99.8%. The positive predictive value—a measure of the accuracy of rule-in—was 74.5%. Using the strategy, physicians ruled out NSTEMI in 57% and ruled it in in 18%.

Among the 3,500 patients with serial troponin I measurements, the negative predictive value was 99.7% and the positive predictive value was 62.3%. The approach led to 44% of patients being ruled out and 23% being ruled in.

Based on either type of troponin assay, 0.1% of patients in the ruled out groups eventually received an NSTEMI diagnosis.

The algorithm performed comparably across subgroups. For the early presenters in particular, the approach ruled out 59% of patients with troponin T measurements and 46% of those with troponin I measurements. Various measures of performance “clearly reject the hypothesis of a substantially lower performance in early presenters,” the authors say.

Use of the 0/1-hour algorithm seemed to provide prognostic information as well, discriminating risks of all-cause mortality and MACE at 30 days and 1 year between the rule-out, observation, and rule-in groups.

Coming to America? Not So Fast

In an accompanying editorial, Joseph Alpert, MD (University of Arizona College of Medicine, Tucson, AZ), and Allan Jaffe, MD (Mayo Clinic, Rochester, MN), highlight some concerns with the 0/1-hour approach, with some overlapping the issues raised by Peacock.

They raise questions about whether the metrics for troponin I used in the study are optimal, about the lack of information on care received after the patients left the ED, about the types of patients included, about the timing between symptom onset and the first troponin measurement, and about assay characteristics.

“The current study and other clinical investigations support the concept that many patients, particularly low-risk individuals who present at least 3 hours after the onset of symptoms, can have acute MI excluded with high-sensitivity troponin testing, assuming [US] Food and Drug Administration approval of the values used extensively in Europe and in this evaluation,” they write. “Even then, however, caution is advised when dealing with patient subsets such as the elderly, the critically ill, individuals with end-stage renal disease, those who present with atypical symptoms, and perhaps also those who present very early after the onset of symptoms.”

Speaking with TCTMD, Alpert indicated that a 0/1-hour strategy might not work at most US centers.

“I think if it’s a younger or middle-aged person who has specific chest pain, that’s why they came to the emergency room, they don’t have kidney failure, there’s not a whole lot of other illnesses complicating it, probably 1 hour is fine,” he said, noting, however, that US hospitals tend to see sicker patients—those with kidney failure, for example—when compared with their European counterparts. “So 1 hour isn’t going to work for those patients. Probably 3 hours will work. It’s going to be a judgment call in US emergency rooms.”

Even so, Alpert said, having high-sensitivity troponin assays available is “going to help us to send more people home from the emergency room than we were able to send in the past.”

For his part, Peacock said that he loves the idea of the 0/1-hour algorithm but that it must be evaluated in US centers using cut points specific to Americans—which are higher than those employed in Europe—before trying to roll it out in practice.

“Until this is validated . . . we do all these research things at our patients’ risk,” he said.

US physicians will soon have more options when choosing high-sensitivity troponin assays. The FDA recently approved two high-sensitivity troponin I assays for use with the Atellica IM and ADVIA Centaur XP/XPT in vitro diagnostic analyzers from Siemens Healthineers.