This article is more than 5 years old. Information may no longer be current.
Q&A: High-sensitivity troponin assays may help diagnose more than MI
Click Here to Manage Email Alerts
Cardiac troponin assays have been used clinically for approximately 30 years by cardiologists and other health care practitioners to diagnose MI. High-sensitivity troponin assays, used for approximately 10 years in Europe, have just started to make their way into clinical practice in the United States. They offer clinicians a faster and more precise way to rule in or rule out MI than previous assays, but questions surrounding their implementation remain.
Cardiology Today spoke to two experts on cardiac troponin assays, Robert H. Christenson, PhD, DABCC, FACB, professor of pathology and professor of medical and research technology at the University of Maryland School of Medicine and past president of the American Association for Clinical Chemistry, and Cardiology Today Editorial Board Member Allan S. Jaffe, MD, professor of medicine and professor of laboratory medicine and pathology and chair of the division of clinical core laboratory services at Mayo Clinic Rochester, for insight on the integration of these tools into clinical practice and how they can be used to help with management of conditions other than MI.
Question: Could you explain how high-sensitivity cardiac troponin assays are different from the ones that cardiologists and emergency care physicians have traditionally used?
Robert Christenson: It is important to realize what the high-sensitivity assays will do and what they will not do relative to earlier-generation tests. The definition of “high sensitivity” is analytical but the benefits will be clinical. The target analyte of both the earlier generation and the modern high-sensitivity assays is the same, ie, cardiac troponin T or I. One of the benefits of the high-sensitivity assays is we have some benchmark metrics, such as they are, to define at least the analytic quality of the assays.
There are two criteria in the definition of high-sensitivity cardiac troponin assays. The first criterion involves precision of the measurements; high-sensitivity assays must have precision, represented by a total coefficient of variation of 10% at the 99th percentile upper reference limits of healthy men and women. This benchmark translates to assays that have measurement uncertainty that is superior to earlier-generation cardiac troponin tests. The second criterion specifies that high-sensitivity tests must be able to detect cardiac troponin at very low concentrations. High-sensitivity assays must be able to reliably detect cardiac troponin in 50% of healthy men and healthy women.
The highest-quality cardiac troponin tests available in the U.S. before the recent high-sensitivity era are termed “contemporary cardiac troponin assays.” At the universal cut point of the 99th percentile of a healthy population of men and women, it was possible to reliably detect the upper 10% of these values below the upper reference limit. With contemporary assays the field were unable to distinguish between subjects with very low cardiac troponin concentrations, from individuals with modest or moderate values that were within the so-called normal reference interval. As a result, dynamic temporal changes in cardiac troponin concentrations consistent with acute injury went undetected with contemporary assays. On the other hand, high-sensitivity assays have the ability to quantify cardiac troponin well below the 99th percentile upper reference limit, so it is not possible to detect and monitor dynamic changes indicating myocardial injury much earlier. The innovation of greater precision at low cardiac troponin concentrations with high-sensitivity assays allows detection of myocardial injury well below the 99th percentile upper reference limit. That advantage is, in part, what drives many of the clinical advantages of high-sensitivity cardiac troponin.
The superior precision and limit of detection implicit in high-sensitivity cardiac troponin assays have led to the possible use of different, more appropriate 99th percentile upper reference limits for men and women. The use of sex-specific upper reference limits may aid with earlier detection of cardiac injury and MI in women, which has been an underserved population in terms of detection and treatment of CAD.
Q: Now that we’ve had some time in clinical practice with these newer assays, what has the effect been on patients? Have we really seen an impact on clinical practice and treatment of patients?
Allan Jaffe: First of all, there is a very robust experience in Europe with the first high-sensitivity troponin assays. They’ve been out there since 2010 and there is a fairly well-developed idea about how to use them. Some of that is unique to the European experience that will need to be adapted in the United States. And some of that is going to be relevant and informative.
The first thing that can be said is that in the right circumstances, one can exclude MI more rapidly than has been the case heretofore. That can be done depending upon the assay and the metrics that are used in the United States — and there are some technical issues there — by using a value that is below the limit of detection of the assay. That assumes that the FDA will approve these assays to be reported down to that level. There may need to be some adaptation of that if the agency does not.
Nonetheless, a value below the limit of detection of the assay in the appropriate patient does work as a “single sample rule out” of MI. Even if that approach is not employed, the ability to look for changes within several hours will facilitate more expeditious exclusion of MI. Most of the protocols being proposed now are 2 or 3 hours rather than the 6-hour ones we previously used. In Europe, there is an affection for a 1-hour protocol based on the thought that the assays are sufficiently sensitive and precise so that one can see changes early. I have some concerns about that, but whether it is a 1-, 2- or 3-hour approach, one of the key initial benefits of these assays will be to exclude MI in appropriate patients more rapidly. Like anything else, there are some technical issues having to do with that. There are some patients who can come in late, and one needs to be careful because in those patients, seeing changes may take more than a few hours, as the downslope of the time-concentration curve is slower than the upslope. There also are some patients who come in very early who may be missed by only a single sample rule out or very short periods of sampling. However, by and large, high sensitivity has, in Europe and thus far in the United States, facilitated more rapid evaluation of these patients.
On the rule-in side, for patients with chest pain, one has a much more rapid triage as well, because one can develop a variety of metrics to evaluate these patients that make the determination that something acute has occurred, far more rapidly. This can be done either by using values that are very high because they are usually more apt to be MIs when they have very large values, or by looking over time at the change in values in a high pre-test probability patient. So, both from the rule-in side and from the rule-out side, these assays will facilitate the evaluation of patients in the ED, and allow for more rapid triage either to home or to the hospital.
Q: Do these assays have other applications of interest?
Jaffe: These assays are not just for cardiologists. In the real world, the ED is very much involved in having to interpret these values for the triage of possible MI. Labs have a role in facilitating this that is terribly important. But there also is a burgeoning area in postsurgical practice where myocardial injury is very common and highly prognostic. Hospitalists are going to use these assays in the inpatient settings for those who are critically ill and eventually internists and general practitioners will use them as well as risk markers for CVD.
What we need to do is to educate cardiologists so that they can interact appropriately with all of these other groups and serve as trainers and to be able to handle questions to facilitate everybody’s use of these assays.
Christenson: We also have to keep in mind that there still are some issues where high-sensitivity assays will have limitations. ne these is mitigation of interferences by, for example, heterophile antibodies. Also, there is a large body of literature about autoantibodies against cardiac troponin. In other words, some individuals produce antibodies against cardiac troponin or that cross-react with cardiac troponin to interfere with the assays. Hemolysis, the breakdown of red blood cells either in vitro or in vivo, can interfere with many cardiac troponin assays. In addition, the coupling of biotin with streptavidin, one of nature’s strongest bonding reactions, is utilized in a number of immunoassays including many high-sensitive cardiac troponin tests. The potential issue is that high doses of biotin, also termed vitamin B7, are administered therapeutically to treat, for example, patients with multiple sclerosis. Also, biotin is sold as a beauty-aid supplement. Laboratorians, cardiologists and other specialties must work together to educate users of high-sensitivity cardiac troponin assays about these and other potential interferences.
Further, a large proportion of patients with chronic or ‘smoldering’ myocardial injury from, for example, renal failure, have cardiac troponin concentrations that exceed the 99th percentile upper reference limit and can be a false positive for MI. Timing of both symptoms onset and serial sampling are important for this interpretation, as is emphasis on dynamic changes in high-sensitivity cardiac troponin concentrations that are consistent with acute myocardial injury and help avoid such false positives.
Jaffe: Another application is that if one wants to improve the ability to predict which patients with atrial fibrillation are at risk for emboli-related mortality and even bleeding, it turns out that high-sensitivity troponin and some additional biomarkers are extremely helpful in improving the predictive accuracy of the CHA2DS2-VASc score. So that, this is going to be, in the future, an important augmentation of their use.
But there also are a variety of implications for when you have elevations in patients who don’t seem to have disease. We still have to work out all the specific details of each scenario, but these patients are at increased risk for CV events. These are the sorts of applications that, in the long run, are very likely to have a major impact on facilitation of care and will be important to cardiologists to know and understand.
Christenson: As mentioned previously, earlier-generation assays were not sensitive enough for detection of a subtle increase in cardiac troponin in patients who were asymptomatic. High-sensitivity assays has opened this whole field of study. There’s a lot of very interesting research going on in that area of study currently.
Jaffe: Another potential application is in pulmonary embolism. There have been trials in Europe asking whether or not we can use high-sensitivity troponin as a surrogate for patients who are sufficiently ill to warrant thrombolytic therapy or more aggressive therapy. And thus far, they’ve been equivocal.
What is terribly important for cardiologists to understand is that anything that hurts the heart is going to cause an increase in troponin. Not all elevations are going to be due to MI. And pulmonary embolism is just one example. Sepsis is another. There are a lot of reasons for myocardial injury. And what cardiologists are going to need to do with along with their colleagues is to make clinical distinctions when one sees elevations in troponin.
And there are two components that can help one think about this as an issue. The critical one is that elevations may be present because of, for example, structural heart disease, which is quite common, and can lead to values being elevated. But if the values are not changing, the explanation is most likely some sort of chronic underlying disease. It may lead to an adverse prognosis and may need to be taken into account as part of ongoing clinical care, but it is not an acute change. On the other hand, a changing pattern of values suggests there is an acute component. If that injury is due to ischemic heart disease, then they have MI. However, a changing pattern is not specific to MI. It could be due to myocarditis or pulmonary embolism. Thus, clinicians need to have a broader purview and thought process than just, “Elevated troponin is a heart attack.” That mindset will no longer serve cardiologists well.
Christenson: To put it another way, all MIs have increased cardiac troponin, but not all elevations in cardiac troponin are MI.
Q: Will the results of the HIGH-STEACS trial impact the use of the high-sensitivity assays around the world?
Jaffe: That was an interesting trial done by the Edinburgh group, which has been very much involved in use of a particular assay (Architect Stat, Abbott) for many years. One thing they had shown previously was that as one used more and more sensitive assays, one improved care, and that this increased sensitivity led to better outcomes. So, the HIGH-STEACS presentation at the European Society of Cardiology Congress was a little bit disappointing to many of us because it didn’t show that there were better outcomes for those patients with chest discomfort evaluated with a high-sensitivity assay compared with those evaluated with a more standard non-high-sensitivity assay. Interestingly, there were many more elevations with the high-sensitivity assay, especially in women.
However, it appeared that when clinicians saw an elevated troponin not clearly due to MI, they didn’t know how to treat those patients or even how to evaluate them, so often no treatment was rendered. If one ignores these more frequent elevations and does not treat the patients, it is not surprising there would not be differences in outcomes. We cardiologists need to take all elevations seriously and to initiate therapy for their specific etiology, which means we have to look into those etiologies in a little more depth.
Although we were disappointed at those results, they also show a potential path forward for improving the use of these tests and making sure that we link these diagnoses to some sort of therapeutic option. If it’s pulmonary embolism, you’ve got to make that diagnosis. If it’s sepsis, you have to ask once the sepsis resolves, what comorbidity might there be in the heart. There are a wide variety of clinical circumstances, but considering myocardial injury as an important diagnosis is part of the emphasis of the new universal definition of MI released in 2018 at the ESC Congress.
Q: What knowledge gaps in this area remain that you ’ d like to see further addressed?
Christenson: We are going to have to be much more precise in the diagnosis of cardiac conditions, particularly in a setting of comorbidities. Can we recognize important changes in cardiac troponin within the normal reference interval that are harbingers of future disease states such as HF? Because of the biological stability of cardiac troponin, when you see an increase, you have to think about myocardial injury, whether it’s related oncological, renal, diabetes or cardiac conditions. Combining cardiac troponin with other biomarkers may also be beneficial in public health.
Jaffe: It’s important for U.S. clinicians to know that most of the data that comes from Europe is predicated on patients who presented with chest pain. The primary question to be addressed was, “Did this patient have a heart attack?” If a patient was more complicated, for example, critically ill with end-stage renal disease, or didn’t have typical symptoms as sometimes the elderly and women may not have, then they generally are not included in European studies such as HIGH-STEACS.
That reflects clinical practice in Europe. They have a heterogeneity of systems, but many of them rely on separating out the primarily rule-out/rule-in patients from those who are critically ill or who present atypically. That is not generally the way in which things go in the United States. If we try to extrapolate everything from Europe, then we’re going to leave a very large gap in terms of the evaluation of the large number of patients who present atypically, such as women and the elderly and the critically ill who often have troponins evaluated in the ED. Because of those exclusions, European studies have a mean age of about 62 years, whereas many American studies have a mean age into the 70s.
When Mayo Clinic implemented the high-sensitivity troponin assay, we decided to make an attempt at what it would take to make our protocol applicable to everyone, as opposed to suggesting that we had to somehow cull out only the rule-in/rule-out subset. We were concerned that if we did not include everyone, we would need multiple sets of criteria. The exact protocols to do that will vary by assay, but we felt that having consistent criteria for everyone would eventually improve our practices.
It should also be appreciated that the majority of the European data, because troponin T was available for use in Europe in 2010, is predicated on troponin T as the gold standard. And that may or may not be the best way to proceed. Perhaps we will come out with different algorithms with some of the more recently approved assays.
Christenson: We need to do more investigation in asymptomatic populations, most of which will initially have cardiac troponin values below the 99th percentile upper reference limit. Use of imaging techniques such as late gadolinium enhancement may provide important insight into structural changes in the heart, which may or may not be related to ischemia. Those changes in the heart reflect injury, and cardiac troponin may represent an early ‘signature’ of disease. Early studies indicate that cardiac troponin appears to be a biological signal for HF that occurs before functional changes as measured by an echocardiogram are apparent. If the echo indicates that that function is still within the normal range, cardiac troponin may provide an early biological signal that could allow early and inexpensive screening of chronic cardiac injury that may be the target of therapy.
Jaffe: In the future, we need to define those protocols because that will be great value. But I’m not sure we’re ready to do that just yet. We are secure in the prognostic value of high-sensitivity increases in groups of patients, but we still need to define how much difference is needed in individual patients before there is a signal that is actionable. There are often only minor differences in cohort studies between those who are at risk and those who are not, so knowing in any given patient that those differences are even analytically real can be a challenge. So, I would argue that it’s going to take us some time to develop specific metrics for those sorts of issues. But we will, and when we do, using high-sensitivity troponin for many chronic issues will be of benefit. The emphasis today in implementation in the United States has to do with its use in diagnosing MI, partially because these other applications are going to take some time to develop and partially because it’s such a critical issue for so many patients.
Christenson: There’s wisdom in this logic. The emphasis and focus must be MI first.
Jaffe: There are not yet a lot of data to help us make these assays even more valuable or even more applicable to the United States environment.
Q: What other challenges remain?
Jaffe: There is a lot of hype in regard to the use of these assays, in particular to facilitate moving people out of the emergency room. And a lot of that’s good. However, there are, as with everything else, details that matter. So, it can turn out that patients who present very early after the onset of an acute event, may not fit the rubric that has been developed, because there are not a huge number of early patients reported in the European experience.
Secondly, the patients that come in late, after the time where the rise might have occurred, can be on the downslope of the so-called “time concentration curve.” That is to say, they’ve already released, and now that they’re on the downslope, and it may take longer to see a delta. So, one needs to use clinical caution and start initiating these protocols and learn. We have found, at least at Mayo Clinic, that we have a fairly substantial proportion of patients with MI, who do not have a delta at the time we see them, because they come in late. These are very important issues for clinicians to keep in mind. – by Erik Swain
Reference:
Shah ASV, et al. Lancet. 2018;doi:10.1016/S0140-6736(18)31923-8.
For more information:
Robert H. Christenson, PhD, DABCC, FACB , can be reached at rhchristenson@umm.edu.
Allan S. Jaffe, MD, can be reached at jaffe.allan@mayo.edu.
Disclosures: Christenson reports he consults for Beckman Coulter, Quidel and Siemens. Jaffe reports he consults for all companies that make troponin assays.