High-sensitivity troponin assays present opportunities, challenges for MI diagnosis

In January, the FDA gave clearance to a next-generation troponin T assay, the first high-sensitivity troponin assay to be approved for use in the United States. Although such technology has been in use in the rest of the world for several years, U.S. clinicians are just beginning to understand how it can affect patients presenting with chest pain who may have MI.

There are myriad implications to the greater sensitivity of next-generation assays such as the one approved earlier this year (Elecsys Troponin T Gen 5 STAT, Roche Diagnostics). Experts interviewed by Cardiology Today said the use of these tests will have a large positive effect on how patients with chest pain are handled in the ED — both in enabling patients experiencing an MI to be more quickly identified and treated, and in enabling doctors to quickly rule out the possibility of MI in a large percentage of patients. Furthermore, new research suggests that detecting small troponin elevations will lead to new clinical benefits in preventive care and management of chronic disease.

At the same time, those interviewed said it is imperative that clinicians are educated on how to appropriately interpret and respond to the results of this test. It will detect elevations of troponin that may be etiologically unrelated to MI, so clinicians must be prepared to think about troponin in new ways.

“It changes the entire paradigm by which we evaluate patients with chest pain,” Venu Menon, MD, FACC, director of the coronary care unit at Cleveland Clinic, told Cardiology Today.

Credit: Shawn Rocco/Duke Health; printed with permission.

Patients confirmed by the assay to have MI could benefit greatly, L. Kristin Newby, MD, MHS, professor of medicine at Duke University Medical Center and a Cardiology Today Editorial Board member, said in an interview.

“It tells us that there’s been myocardial injury, and that’s prognostically important,” Newby said. “Those people, relative to someone with the same condition who didn’t have a troponin elevation, do worse.”

In April, a meta-analysis published in Annals of Internal Medicine demonstrated that a single high-sensitivity cardiac troponin T concentration, in combination with a nonischemic ECG “may successfully rule out [acute MI] in patients presenting to EDs with possible emergency [ACS].”

However, Allan S. Jaffe, MD, FACC, FAHA, FESC, professor of medicine, professor of laboratory medicine and pathology, and chair of the division of clinical core laboratory services at Mayo Clinic and a member of the Cardiology Today Editorial Board, said the FDA has mandated that only troponin T values of 6 ng/L or greater detected by the Roche assay can be reported, which may not be low enough to rule out MI based on one blood draw plus a normal ECG.

“Maybe 6 ng/L is a value that will still be low enough, but that is unclear at present. So this is a strategy that, unfortunately, we [in the United States] will not be able to employ with this assay at present,” Jaffe said.

Technological advancements

Troponin T and troponin I are reliable biomarkers of MI because they are released into the blood when cardiac muscle dies. After MI, troponin levels begin to rise almost immediately and remain elevated for up to 2 weeks. In 1999, the definition of MI was rewritten to name cardiac troponin as the preferred biochemical marker for detecting MI and to specify diagnostic criteria, including detection of elevated troponin.

High-sensitivity cardiac troponin (hs-cTn) assays measure the same isoforms of troponin as conventional assays but can quantify concentrations that were previously undetectable. “They allow us to see a whole range of numbers we never could see before,” Frank Peacock, MD, FACEP, emergency medicine research director at Baylor College of Medicine, Houston, told Cardiology Today.

Using hs-cTn assays, the possibility of MI can be more quickly ruled out for a large percentage of patients who present at the ED with chest pain. If a patient’s troponin level is normal upon arriving and remains low after 1 or 3 hours, depending on the algorithm being used, clinicians can confidently rule out MI, as MI would result in an increase in troponin detectable by hs-cTn assays within this time frame.

A 2015 analysis of the ongoing APACE study found that nearly 60% of patients who presented at the ED with suspected acute MI were ruled out, with a 99.9% negative predictive value, based on a hs-cTnT 1-hour algorithm. However, in an editorial in Circulation, Jaffe and Harvey D. White, MB, ChB, DSc, from Green Lane Cardiovascular Service, Auckland City Hospital, New Zealand, expressed concern that some of the cutoff values suggested may not work well, especially in patients who present early after the onset of chest discomfort.

The Annals of Internal Medicine meta-analysis, led by John W. Pickering, PhD, from Christchurch Hospital, New Zealand, Jaffe, Peacock and colleagues, focused on 9,241 patients from 11 cohort studies. Overall, 30.6% of hs-cTnT assays were classified as low-risk, and only 0.5% of patients had acute MI. None of those patients died. Sensitivity for risk classification for acute MI ranged from 87.5% to 100% (pooled estimate, 98.7%; 95% CI, 96.6-99.5) and sensitivity for 30-day major adverse CV events ranged from 87.9% to 100% (pooled estimate, 98%; 95% CI, 94.7-99.3).

The meta-analysis authors noted that while the results suggest acute MI may be ruled out in many patients after a single blood draw, the strategy should not be implemented without additional clinical assessment or in patients presenting at less than 3 hours of symptom onset.

Peacock told Cardiology Today that while the data suggest that “people with really low troponins are probably fine,” the findings are only valid in patients at 3 hours of symptom onset or longer.

“If you start applying this [strategy] to people who have had 2 hours of chest pain, it will not work and you’ll kill a few people,” he said.

It is too soon to know which diagnostic algorithm will be widely adopted in U.S. hospitals, but experts agreed that, whether a 1-hour or a 3-hour algorithm is applied, the new assay will enable both greater speed and certainty in ruling out MI.

“When this test is negative, it truly is negative and you don’t have to do any additional workup to rule out MI. When the old troponin [test] was negative, you were never quite sure,” Peacock said.

For many patients experiencing MI, the hs-cTn assays also enable greater efficiency on the rule-in side. With conventional assays, a patient might have to wait 6 hours or longer for the assay to detect abnormal levels of troponin. With the new assay, an estimated 15% to 20% of patients will, within 1 or 3 hours, exhibit troponin values high enough and rising quickly enough to be admitted for MI.

“The advantage of that is that we can diagnose patients with MI earlier,” Peacock said. “Before, we had to wait until the levels got above what our test could detect.”

Newby said faster rule-out and rule-in times have significant implications for ED crowding and patient flow through the ED. Most patients will be discharged sooner, and those who are having an MI can be admitted more quickly. “The time savings will be huge,” she said.

Furthermore, with the use of hs-cTn assays, clinicians may be able to detect some MIs that are fully undetected by conventional assays. “Many patients presented in the past with what we thought was unstable angina. Now we’re picking these up because there actually was a small amount of ischemic myocardial injury,” Joseph S. Alpert, MD, professor of medicine at the University of Arizona College of Medicine and a Cardiology Today Editorial Board member, said in an interview.

Although these increases are likely to be modest, Newby said, there is value in detecting these small MIs. “We know, from transitioning from [the CPK-MB test] to troponin many years ago and with each increasingly sensitive generation of troponin assay, that those individuals who are picked up as an MI who wouldn’t have been with a prior assay do better when they are identified because they are getting recommended therapy where potentially they wouldn’t have before.”

The gray zone

Along with faster rule-in and rule-out times, however, comes a clinical challenge.

“When all is said and done, that leaves about 25% of people in the gray zone. They have a troponin that’s elevated, but it’s not elevated enough to say that it’s an MI. They will need the workup, and the key here is getting the right workup for the patient,” Peacock said.

Experts told Cardiology Today that determining the proper course of action for patients in this gray area requires thoughtful attention to multiple factors.

Newby noted that delta values will play a particularly important role, both because patients’ normal levels of troponin vary and because unchanging elevations likely signify problems other than MI. “We’re going to have to encourage people to move away from simply thinking about a threshold value and putting it in the context of the delta, which may become increasingly important for teasing out MI from non-MI troponin elevations,” she said.

“The situation we are most concerned about is the one in which a patient presents immediately after having an acute chest pain episode,” Menon said. Those patients might not yet have a troponin level high enough to be actionable but may be experiencing an acute MI.

“One in five of those patients [in the gray zone] will be having a real injury from an acute ischemic event,” Menon told Cardiology Today. “The challenge for the ED physician in our legal medical environment will be to figure out what kind of tests we need to do to be comfortable sending home the four patients who don’t have an ACS while still recognizing the one who truly does.”

Non-MI troponin elevations

The April 2017 Annals of Internal Medicine meta-analysis prompted discussion over whether results from hs-cTn assays would lead to excessive downstream testing and unnecessary interventions.

Jaffe said clinicians need to understand that “in the absence of signs or symptoms, clinicians should not pursue evaluations for ischemia. The idea that an elevated [cardiac troponin] value has to mean ischemia is not really warranted by the clinical literature.”

According to Peacock, it is imperative to think of troponin as a marker for more than MI, which represents a shift in perspective for many doctors. Historically, troponin assays have effectively functioned as an MI test because an MI is one of the few things that cause troponin elevations to reach the levels detectable by conventional assays. However, thinking of the new assay as a “test” for MI is extremely problematic.

“It is not an MI test,” Peacock said. “It is a ‘part of your heart died’ test. And yes, MI can kill part of your heart, but there are at least a couple hundred other things that are also associated with myocardial cell death,” including renal failure, sepsis, pulmonary embolism and stroke.

Identifying the reasons why troponin is elevated and, if needed, admitting those patients will improve their care and outcomes, he said.

Jaffe said the efficiencies enabled by hs-cTn assays will result in fewer unnecessary hospital admissions, but there are two groups for whom admissions will appropriately rise.

“One group is people who have CVD — HF, for example. Those individuals are at increased risk and troponin helps you know that, so it should be helpful in taking care of those patients,” said Jaffe. “There will be another group of patients who don’t have CVD and who have elevations of troponin. They may have structural heart disease, or they may have underlying [disease] like pneumonia, which directly or indirectly causes cardiac stress and thus an elevated troponin level.”

These are the kinds of patients, according to Jaffe, whose treatment must be carefully considered. Some clinicians fear that all patients with elevated troponin levels will be admitted to cardiology, overwhelming the cardiology department and preventing optimal care of their primary problem. He said it is “a possibility” in the absence of appropriate planning and coordination. This will not happen if “people meet, get together, and make some plans about how one is going to deal with these patients.”

Ultimately, the shift in perspective and clinical practice required by hs-cTn assays requires education, time and commitment.

“ED physicians, cardiologists, laboratorians, internists — we all need to be working together to come up with our protocols and our implementation plans. We have time to do that now and get that all in order,” Newby said.

According to Peacock, the companies such as Roche that manufacture these tests also bear responsibility. “The company has a huge role to play in educating physicians,” he said. “Part of the job of anyone who manufactures a high-sensitivity troponin assay is to explain to physicians that this test is a cardiac injury test — that it indicates that part of the cells of the heart have died, and that you have to do the right thing for that, which doesn’t necessarily mean going straight to the cath lab.”

Transition to a new assay

Newby said the adoption of the new Roche assay will certainly come with transitional challenges: “It’s hard to argue with moving to a higher-quality assay. The opportunities far outweigh the downsides.”

U.S. clinicians may be able to learn from the experience of others around the world, Menon said.

“There is nothing new about this test,” he said. “It is now part of the European Society of Cardiology guidelines as the methodology by which acute chest pain is evaluated in the ED. So, in many ways, we are behind the rest of the world in implementing this technology.”

As doctors and hospitals transition to the Roche assay, experts agreed that some additional logistical challenges are inevitable.

“Doctors have a history with the old troponin — we understand it, we’ve used it for a long time, we’re comfortable with it and we know what to do to get good results,” Peacock said. “When you roll out the new one, that experience will not exist and doctors won’t know what to do.”

For example, troponin values are represented in the new test using different units. Whereas the existing assay produces a value such as 0.05 (ng/mL), values from the new test will be whole numbers (ng/L). This, Alpert said, is to minimize confusion that could arise from the number of zeroes that come with the far smaller troponin concentrations being represented.

Furthermore, new cutoff values must be established and validated for the U.S. ED setting and the U.S. population. “The company will give you values,” Alpert said, “but you probably need to reproduce that in the set of norms at your own hospital.”

Unlike in other countries that already use the Roche test, the FDA approval in the United States specifies different cutoff values for men and women. Based on the 99th percentile upper reference limit, 19 ng/L is identified as the cutoff value overall, with sex-specific cutoff values defined as 14 ng/L for women and 22 ng/L for men.

The values approved for the Roche assay, Jaffe noted, are considerably higher than they have been in most parts of the world, which he said is problematic because much of the data generated about the assay in other countries may not be usable if one adheres to those values.

Newby said she expects early adopting institutions to play a major role in establishing and validating algorithms and cutoff points for the United States. “Once you have a few sentinel early adopters, their protocols and their experiences translate much more easily to hospitals that are a little less eager to be the first one out there,” she said. “Those places will also be meticulous as far as tracking outcomes [regarding] patients, cost and time in the ED.”

Future applications

In the future, some of the most significant changes enabled by hs-cTn may relate not to MI diagnosis but to treatment of chronic disease and preventive care, according to Jaffe.

“Clinicians have for years thought about troponin as an acute marker only,” he said. Jaffe noted that the role of troponin as a prognostic marker is beginning to be better understood, and the clinical implications for using hs-cTn are significant.

“One could begin to look at patients who are at risk, for primary prevention or secondary prevention, and by scrutinizing their troponin values predict who is at increased risk and who is at reduced risk,” Jaffe said. “There are robust data both in terms of primary prevention and for patients who have known CVD, for whom the cutoff values and treatment algorithms are going to be different, but for whom there will be tremendous benefit.”

For example, analysis of data from the WOSCOPS study found that cardiac troponin I levels independently predicted CHD events among men with hypercholesterolemia who had no history of MI. Further, Jaffe said, there are multiple studies showing that both troponin T and troponin I can be used to predict which patients who have atrial fibrillation are at highest risk for stroke, death and bleeding.

Jaffe told Cardiology Today that it will take time for these new uses of troponin to be fully explored, but doctors should not lose sight of the test’s broader long-term potential while sorting through the short-term challenges in learning how to use hs-cTn to aid in MI diagnoses.

“Some of the greatest benefits of detecting low troponin elevations are still being discovered,” he said. – by Sarah Kennedy

• References:
• Christenson E, Christenson RH. Ann Lab Med. 2013;doi:10.3343/alm.2013.33.5.309
• FDA 510(k) Review: Decision Summary. https://www.accessdata.fda.gov/cdrh_docs/reviews/K162895.pdf. Accessed March 30, 2017.
• Ford I, et al. J Am Coll Cardiol. 2016;doi:10.1016/j.jacc.2016.10.020.
• Jaffe AS, White H. Circulation. 2016;doi:10.1161/CIRCULATIONAHA.116.024687.
• Pickering JW, et al. Ann Intern Med. 2017;doi:10.7326/M16-2562.
• Reichlin T, et al. CMAJ. 2015;doi:10.1503/cmaj.141349.

• For more information:
• Joseph S. Alpert, MD, can be reached at University of Arizona Health Science Center, Department of Medicine, 1501 N. Campbell Road, Tucson, AZ 85724; email: jalpert@u.arizona.edu.
• Allan S. Jaffe, MD, FAHA, FACC, FESC, can be reached at Division of Cardiology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905; email: jaffe.allan@mayo.edu.
• Venu Menon, MD, FACC, can be reached at Cleveland Clinic, Mail Code J1-5, 9500 Euclid Ave., Cleveland, OH 44195; email: menonv@ccf.org.
• L. Kristin Newby, MD, MHS, can be reached at Duke University Medical Center, DUMC 3213, Durham, NC 27710; email: kristin.newby@duke.edu.
• Frank Peacock, MD, FACEP, can be reached at Emergency Center, Ben Taub Hospital, 1504 Taub Loop, Houston, TX 77030; email: wfpeacoc@bcm.edu.

Disclosures: Alpert and Menon report no relevant financial disclosures. Jaffe and Peacock report consulting for multiple biomarker companies, including Roche Diagnostics. Newby reports consulting for Philips Healthcare and Roche Diagnostics.